chore(deps): Bump oneccl-bind-pt in /backend/python/diffusers

Bumps oneccl-bind-pt from 2.3.100+xpu to 2.6.0+xpu.

---
updated-dependencies:
- dependency-name: oneccl-bind-pt
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>

.bruno/LocalAI Test Requests/Sound Generation/musicgen.bru

@@ -1,23 +0,0 @@
-meta {
-  name: musicgen
-  type: http
-  seq: 1
-}
-
-post {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/v1/sound-generation
-  body: json
-  auth: none
-}
-
-headers {
-  Content-Type: application/json
-}
-
-body:json {
-  {
-      "model_id": "facebook/musicgen-small",
-      "text": "Exciting 80s Newscast Interstitial",
-      "duration_seconds": 8
-  }
-}

.bruno/LocalAI Test Requests/backend monitor/backend monitor.bru

@@ -1,17 +0,0 @@
-meta {
-  name: backend monitor
-  type: http
-  seq: 4
-}
-
-get {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/backend/monitor
-  body: json
-  auth: none
-}
-
-body:json {
-  {
-    "model": "{{DEFAULT_MODEL}}"
-  }
-}

.bruno/LocalAI Test Requests/backend monitor/backend-shutdown.bru

@@ -1,21 +0,0 @@
-meta {
-  name: backend-shutdown
-  type: http
-  seq: 3
-}
-
-post {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/backend/shutdown
-  body: json
-  auth: none
-}
-
-headers {
-  Content-Type: application/json
-}
-
-body:json {
-  {
-      "model": "{{DEFAULT_MODEL}}"
-  }
-}

.bruno/LocalAI Test Requests/bruno.json

@@ -1,5 +0,0 @@
-{
-  "version": "1",
-  "name": "LocalAI Test Requests",
-  "type": "collection"
-}

.bruno/LocalAI Test Requests/environments/localhost.bru

@@ -1,6 +0,0 @@
-vars {
-  HOST: localhost
-  PORT: 8080
-  DEFAULT_MODEL: gpt-3.5-turbo
-  PROTOCOL: http://
-}

.bruno/LocalAI Test Requests/get models list.bru

@@ -1,11 +0,0 @@
-meta {
-  name: get models list
-  type: http
-  seq: 2
-}
-
-get {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/models
-  body: none
-  auth: none
-}

.bruno/LocalAI Test Requests/image generation/Generate image.bru

@@ -1,25 +0,0 @@
-meta {
-  name: Generate image
-  type: http
-  seq: 1
-}
-
-post {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/v1/images/generations
-  body: json
-  auth: none
-}
-
-headers {
-  Content-Type: application/json
-}
-
-body:json {
-  {
-    "prompt": "<positive prompt>|<negative prompt>",
-    "model": "model-name",
-    "step": 51,
-    "size": "1024x1024",
-    "image": ""
-  }
-}

.bruno/LocalAI Test Requests/llm text/-completions.bru

@@ -1,24 +0,0 @@
-meta {
-  name: -completions
-  type: http
-  seq: 4
-}
-
-post {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/completions
-  body: json
-  auth: none
-}
-
-headers {
-  Content-Type: application/json
-}
-
-body:json {
-  {
-      "model": "{{DEFAULT_MODEL}}",
-      "prompt": "function downloadFile(string url, string outputPath) {",
-      "max_tokens": 256,
-      "temperature": 0.5
-  }
-}

.bruno/LocalAI Test Requests/llm text/-edits.bru

@@ -1,23 +0,0 @@
-meta {
-  name: -edits
-  type: http
-  seq: 5
-}
-
-post {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/edits
-  body: json
-  auth: none
-}
-
-headers {
-  Content-Type: application/json
-}
-
-body:json {
-  {
-      "model": "{{DEFAULT_MODEL}}",
-      "input": "What day of the wek is it?",
-      "instruction": "Fix the spelling mistakes"
-  }
-}

.bruno/LocalAI Test Requests/llm text/-embeddings.bru

@@ -1,22 +0,0 @@
-meta {
-  name: -embeddings
-  type: http
-  seq: 6
-}
-
-post {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/embeddings
-  body: json
-  auth: none
-}
-
-headers {
-  Content-Type: application/json
-}
-
-body:json {
-  {
-      "model": "{{DEFAULT_MODEL}}",
-      "input": "A STRANGE GAME.\nTHE ONLY WINNING MOVE IS NOT TO PLAY.\n\nHOW ABOUT A NICE GAME OF CHESS?"
-  }
-}

.bruno/LocalAI Test Requests/llm text/chat/chat completion -simple- 1 message-.bru

@@ -1,30 +0,0 @@
-meta {
-  name: chat completion -simple- 1 message-
-  type: http
-  seq: 4
-}
-
-post {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/chat/completions
-  body: json
-  auth: none
-}
-
-headers {
-  Content-Type: application/json
-}
-
-body:json {
-  {
-    "model": "{{DEFAULT_MODEL}}",
-    "messages": [
-      {
-        "role": "user",
-        "content": "How could one use friction to cook an egg?"
-      }
-    ],
-    "max_tokens": 256,
-    "temperature": 0.2,
-    "grammar": ""
-  }
-}

.bruno/LocalAI Test Requests/llm text/chat/chat-completions -long-.bru

@@ -1,29 +0,0 @@
-meta {
-  name: chat-completions -long-
-  type: http
-  seq: 5
-}
-
-post {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/chat/completions
-  body: json
-  auth: none
-}
-
-headers {
-  Content-Type: application/json
-}
-
-body:json {
-  {
-       "model": "{{DEFAULT_MODEL}}",
-       "messages": [{"role": "system", "content": "You are a helpful, intelligent pirate assistant. Always answer as helpfully as possible, while being a pirate and using pirate language. If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you don't know the answer to a question, please don't share false information."},
-           {"role": "user", "content": "How could one use electricity to cook an egg?"},
-           {"role": "assistant",
-                  "content": "Shiver me timbers! Using electricity to cook an egg? Well, matey, I reckon that be a right curious idea! *adjusts spectacles* Now, I ain't sure if this be possible, but I'll do me best to help ye out.\n\nFirst things first, ye gotta understand that electricity be a powerful force, and it can't just cook an egg on its own. Ye see, electricity be like a mighty wind that can make things happen, but it needs somethin' to work with. So, if ye want to use electricity to cook an egg, ye gotta find a way to harness that power and make it do the cookin'.\n\nNow, I know what ye might be thinkin': \"How do I harness the power of electricity to cook an egg?\" Well, matey, there be a few ways to do it. One way be to use a special device called an \"electric frying pan.\" This be a pan that has a built-in heating element that gets hot when ye plug it into a wall socket. When the element gets hot, ye can crack an egg into the pan and watch as it cook"
-              },
-              {"role": "user", "content": "I don't have one of those, just a raw wire and plenty of power! How do we get it done?"}],
-       "max_tokens": 1024,
-       "temperature": 0.5
-  }
-}

.bruno/LocalAI Test Requests/llm text/chat/chat-completions -stream-.bru

@@ -1,25 +0,0 @@
-meta {
-  name: chat-completions -stream-
-  type: http
-  seq: 6
-}
-
-post {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/chat/completions
-  body: json
-  auth: none
-}
-
-headers {
-  Content-Type: application/json
-}
-
-body:json {
-  {
-       "model": "{{DEFAULT_MODEL}}",
-       "messages": [{"role": "user", "content": "Explain how I can set sail on the ocean using only power generated by seagulls?"}],
-       "max_tokens": 256,
-       "temperature": 0.9,
-       "stream": true
-  }
-}

.bruno/LocalAI Test Requests/model gallery/add model gallery.bru

@@ -1,22 +0,0 @@
-meta {
-  name: add model gallery
-  type: http
-  seq: 10
-}
-
-post {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/models/galleries
-  body: json
-  auth: none
-}
-
-headers {
-  Content-Type: application/json
-}
-
-body:json {
-  {
-      "url": "file:///home/dave/projects/model-gallery/huggingface/TheBloke__CodeLlama-7B-Instruct-GGML.yaml",
-      "name": "test"
-  }
-}

.bruno/LocalAI Test Requests/model gallery/delete model gallery.bru

@@ -1,21 +0,0 @@
-meta {
-  name: delete model gallery
-  type: http
-  seq: 11
-}
-
-delete {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/models/galleries
-  body: json
-  auth: none
-}
-
-headers {
-  Content-Type: application/json
-}
-
-body:json {
-  {
-      "name": "test"
-  }
-}

.bruno/LocalAI Test Requests/model gallery/list MODELS in galleries.bru

@@ -1,11 +0,0 @@
-meta {
-  name: list MODELS in galleries
-  type: http
-  seq: 7
-}
-
-get {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/models/available
-  body: none
-  auth: none
-}

.bruno/LocalAI Test Requests/model gallery/list model GALLERIES.bru

@@ -1,11 +0,0 @@
-meta {
-  name: list model GALLERIES
-  type: http
-  seq: 8
-}
-
-get {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/models/galleries
-  body: none
-  auth: none
-}

.bruno/LocalAI Test Requests/model gallery/model delete.bru

@@ -1,11 +0,0 @@
-meta {
-  name: model delete
-  type: http
-  seq: 7
-}
-
-post {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/models/galleries
-  body: none
-  auth: none
-}

.bruno/LocalAI Test Requests/model gallery/model gallery apply -gist-.bru

@@ -1,21 +0,0 @@
-meta {
-  name: model gallery apply -gist-
-  type: http
-  seq: 12
-}
-
-post {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/models/apply
-  body: json
-  auth: none
-}
-
-headers {
-  Content-Type: application/json
-}
-
-body:json {
-  {
-      "id": "TheBloke__CodeLlama-7B-Instruct-GGML__codellama-7b-instruct.ggmlv3.Q2_K.bin"
-  }
-}

.bruno/LocalAI Test Requests/model gallery/model gallery apply.bru

@@ -1,22 +0,0 @@
-meta {
-  name: model gallery apply
-  type: http
-  seq: 9
-}
-
-post {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/models/apply
-  body: json
-  auth: none
-}
-
-headers {
-  Content-Type: application/json
-}
-
-body:json {
-  {
-      "id": "dave@TheBloke__CodeLlama-7B-Instruct-GGML__codellama-7b-instruct.ggmlv3.Q3_K_S.bin",
-      "name": "codellama7b"
-  }
-}

.bruno/LocalAI Test Requests/transcription/transcribe.bru

@@ -1,16 +0,0 @@
-meta {
-  name: transcribe
-  type: http
-  seq: 1
-}
-
-post {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/v1/audio/transcriptions
-  body: multipartForm
-  auth: none
-}
-
-body:multipart-form {
-  file: @file(transcription/gb1.ogg)
-  model: whisper-1
-}

.bruno/LocalAI Test Requests/tts/-tts.bru

@@ -1,22 +0,0 @@
-meta {
-  name: -tts
-  type: http
-  seq: 2
-}
-
-post {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/tts
-  body: json
-  auth: none
-}
-
-headers {
-  Content-Type: application/json
-}
-
-body:json {
-  {
-      "model": "{{DEFAULT_MODEL}}",
-      "input": "A STRANGE GAME.\nTHE ONLY WINNING MOVE IS NOT TO PLAY.\n\nHOW ABOUT A NICE GAME OF CHESS?"
-  }
-}

.bruno/LocalAI Test Requests/tts/musicgen.bru

@@ -1,23 +0,0 @@
-meta {
-  name: musicgen
-  type: http
-  seq: 2
-}
-
-post {
-  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/tts
-  body: json
-  auth: none
-}
-
-headers {
-  Content-Type: application/json
-}
-
-body:json {
-  {
-      "backend": "transformers",
-      "model": "facebook/musicgen-small",
-      "input": "80s Synths playing Jazz"
-  }
-}

.github/labeler.yml

@@ -1,4 +1,4 @@
-enhancements:
+enhancement:
  - head-branch: ['^feature', 'feature']
 
 dependencies:

.github/workflows/bump_deps.yaml

@@ -9,7 +9,7 @@ jobs:
       fail-fast: false
       matrix:
         include:
-          - repository: "ggerganov/llama.cpp"
+          - repository: "ggml-org/llama.cpp"
             variable: "CPPLLAMA_VERSION"
             branch: "master"
           - repository: "ggerganov/whisper.cpp"

.github/workflows/deploy-explorer.yaml

@@ -33,7 +33,7 @@ jobs:
         run: |
           CGO_ENABLED=0 make build-api
       - name: rm
-        uses: appleboy/ssh-action@v1.2.0
+        uses: appleboy/ssh-action@v1.2.2
         with:
             host: ${{ secrets.EXPLORER_SSH_HOST }}
             username: ${{ secrets.EXPLORER_SSH_USERNAME }}
@@ -53,7 +53,7 @@ jobs:
             rm: true
             target: ./local-ai
       - name: restarting
-        uses: appleboy/ssh-action@v1.2.0
+        uses: appleboy/ssh-action@v1.2.2
         with:
             host: ${{ secrets.EXPLORER_SSH_HOST }}
             username: ${{ secrets.EXPLORER_SSH_USERNAME }}

.github/workflows/generate_grpc_cache.yaml

@@ -2,9 +2,10 @@ name: 'generate and publish GRPC docker caches'
 
 on:
   workflow_dispatch:
-  push:
-    branches:
-      - master
+
+  schedule:
+    # daily at midnight
+    - cron: '0 0 * * *'
 
 concurrency:
   group: grpc-cache-${{ github.head_ref || github.ref }}-${{ github.repository }}
@@ -16,7 +17,7 @@ jobs:
       matrix:
         include:
           - grpc-base-image: ubuntu:22.04
-            runs-on: 'ubuntu-latest'
+            runs-on: 'arc-runner-set'
             platforms: 'linux/amd64,linux/arm64'
     runs-on: ${{matrix.runs-on}}
     steps:

.github/workflows/image_build.yml

@@ -310,6 +310,11 @@ jobs:
           tags: ${{ steps.meta_aio_dockerhub.outputs.tags }}
           labels: ${{ steps.meta_aio_dockerhub.outputs.labels }}
 
+      - name: Cleanup
+        run: |
+          docker builder prune -f
+          docker system prune --force --volumes --all
+
       - name: Latest tag
         # run this on branches, when it is a tag and there is a latest-image defined
         if: github.event_name != 'pull_request' && inputs.latest-image != ''  && github.ref_type == 'tag'

LICENSE

@@ -1,6 +1,6 @@
 MIT License
 
-Copyright (c) 2023-2024 Ettore Di Giacinto (mudler@localai.io)
+Copyright (c) 2023-2025 Ettore Di Giacinto (mudler@localai.io)
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal

Makefile

@@ -6,7 +6,7 @@ BINARY_NAME=local-ai
 DETECT_LIBS?=true
 
 # llama.cpp versions
-CPPLLAMA_VERSION?=d2fe216fb2fb7ca8627618c9ea3a2e7886325780
+CPPLLAMA_VERSION?=1e2f78a00450593e2dfa458796fcdd9987300dfc
 
 # whisper.cpp version
 WHISPER_REPO?=https://github.com/ggerganov/whisper.cpp
@@ -22,7 +22,7 @@ BARKCPP_VERSION?=v1.0.0
 
 # stablediffusion.cpp (ggml)
 STABLEDIFFUSION_GGML_REPO?=https://github.com/leejet/stable-diffusion.cpp
-STABLEDIFFUSION_GGML_VERSION?=d46ed5e184b97c2018dc2e8105925bdb8775e02c
+STABLEDIFFUSION_GGML_VERSION?=19d876ee300a055629926ff836489901f734f2b7
 
 ONNX_VERSION?=1.20.0
 ONNX_ARCH?=x64

README.md

@@ -212,7 +212,7 @@ A huge thank you to our generous sponsors who support this project covering CI e
 
 <p align="center">
   <a href="https://www.spectrocloud.com/" target="blank">
-    <img height="200" src="https://github.com/go-skynet/LocalAI/assets/2420543/68a6f3cb-8a65-4a4d-99b5-6417a8905512">
+    <img height="200" src="https://github.com/user-attachments/assets/72eab1dd-8b93-4fc0-9ade-84db49f24962">
   </a>
   <a href="https://www.premai.io/" target="blank">
     <img height="200" src="https://github.com/mudler/LocalAI/assets/2420543/42e4ca83-661e-4f79-8e46-ae43689683d6"> <br>

aio/cpu/vad.yaml

@@ -0,0 +1,8 @@
+backend: silero-vad
+name: silero-vad
+parameters:
+  model: silero-vad.onnx
+download_files:
+- filename: silero-vad.onnx
+  uri: https://huggingface.co/onnx-community/silero-vad/resolve/main/onnx/model.onnx
+  sha256: a4a068cd6cf1ea8355b84327595838ca748ec29a25bc91fc82e6c299ccdc5808

aio/entrypoint.sh

@@ -129,7 +129,7 @@ detect_gpu
 detect_gpu_size
 
 PROFILE="${PROFILE:-$GPU_SIZE}" # default to cpu
-export MODELS="${MODELS:-/aio/${PROFILE}/embeddings.yaml,/aio/${PROFILE}/rerank.yaml,/aio/${PROFILE}/text-to-speech.yaml,/aio/${PROFILE}/image-gen.yaml,/aio/${PROFILE}/text-to-text.yaml,/aio/${PROFILE}/speech-to-text.yaml,/aio/${PROFILE}/vision.yaml}"
+export MODELS="${MODELS:-/aio/${PROFILE}/embeddings.yaml,/aio/${PROFILE}/rerank.yaml,/aio/${PROFILE}/text-to-speech.yaml,/aio/${PROFILE}/image-gen.yaml,/aio/${PROFILE}/text-to-text.yaml,/aio/${PROFILE}/speech-to-text.yaml,/aio/${PROFILE}/vad.yaml,/aio/${PROFILE}/vision.yaml}"
 
 check_vars
 

aio/gpu-8g/vad.yaml

@@ -0,0 +1,8 @@
+backend: silero-vad
+name: silero-vad
+parameters:
+  model: silero-vad.onnx
+download_files:
+- filename: silero-vad.onnx
+  uri: https://huggingface.co/onnx-community/silero-vad/resolve/main/onnx/model.onnx
+  sha256: a4a068cd6cf1ea8355b84327595838ca748ec29a25bc91fc82e6c299ccdc5808

aio/intel/vad.yaml

@@ -0,0 +1,8 @@
+backend: silero-vad
+name: silero-vad
+parameters:
+  model: silero-vad.onnx
+download_files:
+- filename: silero-vad.onnx
+  uri: https://huggingface.co/onnx-community/silero-vad/resolve/main/onnx/model.onnx
+  sha256: a4a068cd6cf1ea8355b84327595838ca748ec29a25bc91fc82e6c299ccdc5808

backend/backend.proto

@@ -165,7 +165,6 @@ message Reply {
 
 message GrammarTrigger {
   string word = 1;
-  bool at_start = 2; 
 }
 
 message ModelOptions {
@@ -229,6 +228,11 @@ message ModelOptions {
   int32  MaxModelLen = 54;
   int32  TensorParallelSize = 55;
   string LoadFormat = 58;
+  bool   DisableLogStatus = 66;
+  string DType = 67;
+  int32  LimitImagePerPrompt = 68;
+  int32  LimitVideoPerPrompt = 69;
+  int32  LimitAudioPerPrompt = 70;
 
   string MMProj = 41;
 

backend/cpp/llama/grpc-server.cpp

@@ -469,7 +469,7 @@ struct llama_server_context
     bool has_eos_token      = true;
 
     bool grammar_lazy = false;
-    std::vector<common_grammar_trigger> grammar_trigger_words;
+    std::vector<common_grammar_trigger> grammar_triggers;
 
     int32_t n_ctx;  // total context for all clients / slots
 
@@ -709,7 +709,7 @@ struct llama_server_context
         slot->sparams.grammar           = json_value(data, "grammar",           default_sparams.grammar);
         slot->sparams.n_probs           = json_value(data, "n_probs",           default_sparams.n_probs);
         slot->sparams.min_keep          = json_value(data, "min_keep",          default_sparams.min_keep);
-        slot->sparams.grammar_trigger_words = grammar_trigger_words;
+        slot->sparams.grammar_triggers = grammar_triggers;
         slot->sparams.grammar_lazy = grammar_lazy;
 
         if (slot->n_predict > 0 && slot->params.n_predict > slot->n_predict) {
@@ -1155,6 +1155,14 @@ struct llama_server_context
             slot.has_next_token = false;
         }
 
+        if (slot.n_past >= slot.n_ctx) {
+            slot.truncated      = true;
+            slot.stopped_limit = true;
+            slot.has_next_token = false;
+
+            LOG_VERBOSE("stopped due to running out of context capacity", {});
+        }
+
         if (result.tok == llama_vocab_eos(vocab) || llama_vocab_is_eog(vocab, result.tok))
         {
             slot.stopped_eos = true;
@@ -1342,7 +1350,7 @@ struct llama_server_context
         queue_results.send(res);
     }
 
-    void send_embedding(llama_client_slot &slot)
+    void send_embedding(llama_client_slot &slot, const llama_batch & batch)
     {
         task_result res;
         res.id = slot.task_id;
@@ -1364,10 +1372,38 @@ struct llama_server_context
         else
         {
             const float *data = llama_get_embeddings(ctx);
-            std::vector<float> embedding(data, data + n_embd);
+            std::vector<float> embd_res(n_embd, 0.0f);
+            std::vector<std::vector<float>> embedding;
+            for (int i = 0; i < batch.n_tokens; ++i) {
+                if (!batch.logits[i] || batch.seq_id[i][0] != slot.id) {
+                    continue;
+                }
+
+                const float * embd = llama_get_embeddings_seq(ctx, batch.seq_id[i][0]);
+                if (embd == NULL) {
+                    embd = llama_get_embeddings_ith(ctx, i);
+                }
+
+                if (embd == NULL) {
+                    LOG("failed to get embeddings");
+
+                    continue;
+                }
+
+                // normalize only when there is pooling
+                // TODO: configurable
+                if (llama_pooling_type(ctx) != LLAMA_POOLING_TYPE_NONE) {
+                    common_embd_normalize(embd, embd_res.data(), n_embd, 2);
+                    embedding.push_back(embd_res);
+                } else {
+                    embedding.push_back({ embd, embd + n_embd });
+                }
+            }
+
+            // OAI compat
             res.result_json = json
             {
-                {"embedding", embedding },
+                {"embedding", embedding[0] },
             };
         }
         queue_results.send(res);
@@ -1627,17 +1663,17 @@ struct llama_server_context
             {
                 if (slot.is_processing() && system_tokens.size() + slot.cache_tokens.size() >= (size_t) slot.n_ctx)
                 {
+                    // this check is redundant (for good)
+                    // we should never get here, because generation should already stopped in process_token()
+
                     // START LOCALAI changes
                     // Temporary disable context-shifting as it can lead to infinite loops (issue: https://github.com/ggerganov/llama.cpp/issues/3969)
                     // See: https://github.com/mudler/LocalAI/issues/1333
                     // Context is exhausted, release the slot
                     slot.release();
                     send_final_response(slot);
-                    slot.cache_tokens.clear();
-                    slot.n_past = 0;
-                    slot.truncated = false;
-                    slot.has_next_token = true;
-                    LOG("Context exhausted. Slot %d released (%d tokens in cache)\n", slot.id, (int) slot.cache_tokens.size());
+                    slot.has_next_token = false;
+                    LOG_ERROR("context is exhausted, release the slot", {});
 
                     continue;
                     // END LOCALAI changes
@@ -1988,7 +2024,7 @@ struct llama_server_context
                 // prompt evaluated for embedding
                 if (slot.embedding)
                 {
-                    send_embedding(slot);
+                    send_embedding(slot, batch_view);
                     slot.release();
                     slot.i_batch = -1;
                     continue;
@@ -2385,12 +2421,12 @@ static void params_parse(const backend::ModelOptions* request,
         llama.grammar_lazy = true;
         for (int i = 0; i < request->grammartriggers_size(); i++) {
             common_grammar_trigger trigger;
-            trigger.word = request->grammartriggers(i).word();
-            trigger.at_start = request->grammartriggers(i).at_start();
-            llama.grammar_trigger_words.push_back(trigger);
+	    trigger.type = COMMON_GRAMMAR_TRIGGER_TYPE_WORD;
+            trigger.value = request->grammartriggers(i).word();
+	    // trigger.at_start = request->grammartriggers(i).at_start();
+            llama.grammar_triggers.push_back(trigger);
             LOG_INFO("grammar trigger", {
-                { "word", trigger.word },
-                { "at_start", trigger.at_start }
+                { "word", trigger.value },
             });
         }
     }

backend/go/image/stablediffusion-ggml/gosd.cpp

@@ -35,6 +35,8 @@ const char* sample_method_str[] = {
     "ipndm",
     "ipndm_v",
     "lcm",
+    "ddim_trailing",
+    "tcd",
 };
 
 // Names of the sigma schedule overrides, same order as sample_schedule in stable-diffusion.h
@@ -173,6 +175,7 @@ int gen_image(char *text, char *negativeText, int width, int height, int steps,
                             -1, //clip_skip
                             cfg_scale, // sfg_scale
                             3.5f,
+			    0, // eta
                             width,
                             height,
                             sample_method, 

backend/python/coqui/requirements-cpu.txt

@@ -1,4 +1,4 @@
-transformers
+transformers==4.48.3
 accelerate
 torch==2.4.1
 coqui-tts

backend/python/coqui/requirements-cublas11.txt

@@ -1,6 +1,6 @@
 --extra-index-url https://download.pytorch.org/whl/cu118
 torch==2.4.1+cu118
 torchaudio==2.4.1+cu118
-transformers
+transformers==4.48.3
 accelerate
 coqui-tts

backend/python/coqui/requirements-cublas12.txt

@@ -1,5 +1,5 @@
 torch==2.4.1
 torchaudio==2.4.1
-transformers
+transformers==4.48.3
 accelerate
 coqui-tts

backend/python/coqui/requirements-hipblas.txt

@@ -1,6 +1,6 @@
 --extra-index-url https://download.pytorch.org/whl/rocm6.0
 torch==2.4.1+rocm6.0
 torchaudio==2.4.1+rocm6.0
-transformers
+transformers==4.48.3
 accelerate
 coqui-tts

backend/python/coqui/requirements-intel.txt

@@ -5,6 +5,6 @@ torchaudio==2.3.1+cxx11.abi
 oneccl_bind_pt==2.3.100+xpu
 optimum[openvino]
 setuptools
-transformers
+transformers==4.48.3
 accelerate
 coqui-tts

backend/python/diffusers/backend.py

@@ -159,6 +159,18 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
                 torchType = torch.float16
                 variant = "fp16"
 
+            options = request.Options
+
+            # empty dict
+            self.options = {}
+
+            # The options are a list of strings in this form optname:optvalue
+            # We are storing all the options in a dict so we can use it later when
+            # generating the images
+            for opt in options:
+                key, value = opt.split(":")
+                self.options[key] = value
+
             local = False
             modelFile = request.Model
 
@@ -441,6 +453,9 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
         # create a dictionary of parameters by using the keys from EnableParameters and the values from defaults
         kwargs = {key: options.get(key) for key in keys if key in options}
 
+        # populate kwargs from self.options.
+        kwargs.update(self.options)
+
         # Set seed
         if request.seed > 0:
             kwargs["generator"] = torch.Generator(device=self.device).manual_seed(

backend/python/diffusers/requirements-intel.txt

@@ -2,7 +2,7 @@
 intel-extension-for-pytorch==2.3.110+xpu
 torch==2.3.1+cxx11.abi
 torchvision==0.18.1+cxx11.abi
-oneccl_bind_pt==2.3.100+xpu
+oneccl_bind_pt==2.6.0+xpu
 optimum[openvino]
 setuptools
 diffusers

backend/python/vllm/backend.py

@@ -109,6 +109,17 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
             engine_args.swap_space = request.SwapSpace
         if request.MaxModelLen != 0:
             engine_args.max_model_len = request.MaxModelLen
+        if request.DisableLogStatus:
+            engine_args.disable_log_status = request.DisableLogStatus
+        if request.DType != "":
+            engine_args.dtype = request.DType
+        if request.LimitImagePerPrompt != 0 or request.LimitVideoPerPrompt != 0 or request.LimitAudioPerPrompt != 0:
+            # limit-mm-per-prompt defaults to 1 per modality, based on vLLM docs
+            engine_args.limit_mm_per_prompt = {
+                "image": max(request.LimitImagePerPrompt, 1),
+                "video": max(request.LimitVideoPerPrompt, 1),
+                "audio": max(request.LimitAudioPerPrompt, 1)
+            }
 
         try:
             self.llm = AsyncLLMEngine.from_engine_args(engine_args)
@@ -269,7 +280,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
     def load_image(self, image_path: str):
         """
         Load an image from the given file path or base64 encoded data.
-        
+
         Args:
             image_path (str): The path to the image file or base64 encoded data.
 
@@ -288,7 +299,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
     def load_video(self, video_path: str):
         """
         Load a video from the given file path.
-        
+
         Args:
             video_path (str): The path to the image file.
 
@@ -335,4 +346,4 @@ if __name__ == "__main__":
     )
     args = parser.parse_args()
 
-    asyncio.run(serve(args.addr))
+    asyncio.run(serve(args.addr))

core/application/startup.go

@@ -145,13 +145,7 @@ func New(opts ...config.AppOption) (*Application, error) {
 
 	if options.LoadToMemory != nil {
 		for _, m := range options.LoadToMemory {
-			cfg, err := application.BackendLoader().LoadBackendConfigFileByName(m, options.ModelPath,
-				config.LoadOptionDebug(options.Debug),
-				config.LoadOptionThreads(options.Threads),
-				config.LoadOptionContextSize(options.ContextSize),
-				config.LoadOptionF16(options.F16),
-				config.ModelPath(options.ModelPath),
-			)
+			cfg, err := application.BackendLoader().LoadBackendConfigFileByNameDefaultOptions(m, options)
 			if err != nil {
 				return nil, err
 			}

core/backend/llm.go

@@ -33,7 +33,7 @@ type TokenUsage struct {
 	TimingTokenGeneration  float64
 }
 
-func ModelInference(ctx context.Context, s string, messages []schema.Message, images, videos, audios []string, loader *model.ModelLoader, c config.BackendConfig, o *config.ApplicationConfig, tokenCallback func(string, TokenUsage) bool) (func() (LLMResponse, error), error) {
+func ModelInference(ctx context.Context, s string, messages []schema.Message, images, videos, audios []string, loader *model.ModelLoader, c *config.BackendConfig, o *config.ApplicationConfig, tokenCallback func(string, TokenUsage) bool) (func() (LLMResponse, error), error) {
 	modelFile := c.Model
 
 	// Check if the modelFile exists, if it doesn't try to load it from the gallery
@@ -48,7 +48,7 @@ func ModelInference(ctx context.Context, s string, messages []schema.Message, im
 		}
 	}
 
-	opts := ModelOptions(c, o)
+	opts := ModelOptions(*c, o)
 	inferenceModel, err := loader.Load(opts...)
 	if err != nil {
 		return nil, err
@@ -84,7 +84,7 @@ func ModelInference(ctx context.Context, s string, messages []schema.Message, im
 
 	// in GRPC, the backend is supposed to answer to 1 single token if stream is not supported
 	fn := func() (LLMResponse, error) {
-		opts := gRPCPredictOpts(c, loader.ModelPath)
+		opts := gRPCPredictOpts(*c, loader.ModelPath)
 		opts.Prompt = s
 		opts.Messages = protoMessages
 		opts.UseTokenizerTemplate = c.TemplateConfig.UseTokenizerTemplate
@@ -116,6 +116,11 @@ func ModelInference(ctx context.Context, s string, messages []schema.Message, im
 		}
 
 		if tokenCallback != nil {
+
+			if c.TemplateConfig.ReplyPrefix != "" {
+				tokenCallback(c.TemplateConfig.ReplyPrefix, tokenUsage)
+			}
+
 			ss := ""
 
 			var partialRune []byte
@@ -165,8 +170,13 @@ func ModelInference(ctx context.Context, s string, messages []schema.Message, im
 			tokenUsage.TimingTokenGeneration = reply.TimingTokenGeneration
 			tokenUsage.TimingPromptProcessing = reply.TimingPromptProcessing
 
+			response := string(reply.Message)
+			if c.TemplateConfig.ReplyPrefix != "" {
+				response = c.TemplateConfig.ReplyPrefix + response
+			}
+
 			return LLMResponse{
-				Response: string(reply.Message),
+				Response: response,
 				Usage:    tokenUsage,
 			}, err
 		}

core/backend/options.go

@@ -122,7 +122,6 @@ func grpcModelOpts(c config.BackendConfig) *pb.ModelOptions {
 	for _, t := range c.FunctionsConfig.GrammarConfig.GrammarTriggers {
 		triggers = append(triggers, &pb.GrammarTrigger{
 			Word:    t.Word,
-			AtStart: t.AtStart,
 		})
 
 	}
@@ -159,6 +158,12 @@ func grpcModelOpts(c config.BackendConfig) *pb.ModelOptions {
 		SwapSpace:            int32(c.SwapSpace),
 		MaxModelLen:          int32(c.MaxModelLen),
 		TensorParallelSize:   int32(c.TensorParallelSize),
+		DisableLogStatus:     c.DisableLogStatus,
+		DType:                c.DType,
+		// LimitMMPerPrompt vLLM
+		LimitImagePerPrompt:  int32(c.LimitMMPerPrompt.LimitImagePerPrompt),
+		LimitVideoPerPrompt:  int32(c.LimitMMPerPrompt.LimitVideoPerPrompt),
+		LimitAudioPerPrompt:  int32(c.LimitMMPerPrompt.LimitAudioPerPrompt),
 		MMProj:               c.MMProj,
 		FlashAttention:       c.FlashAttention,
 		CacheTypeKey:         c.CacheTypeK,

core/backend/rerank.go

@@ -9,10 +9,10 @@ import (
 	model "github.com/mudler/LocalAI/pkg/model"
 )
 
-func Rerank(modelFile string, request *proto.RerankRequest, loader *model.ModelLoader, appConfig *config.ApplicationConfig, backendConfig config.BackendConfig) (*proto.RerankResult, error) {
-
-	opts := ModelOptions(backendConfig, appConfig, model.WithModel(modelFile))
+func Rerank(request *proto.RerankRequest, loader *model.ModelLoader, appConfig *config.ApplicationConfig, backendConfig config.BackendConfig) (*proto.RerankResult, error) {
+	opts := ModelOptions(backendConfig, appConfig)
 	rerankModel, err := loader.Load(opts...)
+
 	if err != nil {
 		return nil, err
 	}

core/backend/soundgeneration.go

@@ -13,7 +13,6 @@ import (
 )
 
 func SoundGeneration(
-	modelFile string,
 	text string,
 	duration *float32,
 	temperature *float32,
@@ -25,8 +24,9 @@ func SoundGeneration(
 	backendConfig config.BackendConfig,
 ) (string, *proto.Result, error) {
 
-	opts := ModelOptions(backendConfig, appConfig, model.WithModel(modelFile))
+	opts := ModelOptions(backendConfig, appConfig)
 	soundGenModel, err := loader.Load(opts...)
+
 	if err != nil {
 		return "", nil, err
 	}
@@ -44,7 +44,7 @@ func SoundGeneration(
 
 	res, err := soundGenModel.SoundGeneration(context.Background(), &proto.SoundGenerationRequest{
 		Text:        text,
-		Model:       modelFile,
+		Model:       backendConfig.Model,
 		Dst:         filePath,
 		Sample:      doSample,
 		Duration:    duration,

core/backend/tokenize.go

@@ -4,19 +4,17 @@ import (
 	"github.com/mudler/LocalAI/core/config"
 	"github.com/mudler/LocalAI/core/schema"
 	"github.com/mudler/LocalAI/pkg/grpc"
-	model "github.com/mudler/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/pkg/model"
 )
 
 func ModelTokenize(s string, loader *model.ModelLoader, backendConfig config.BackendConfig, appConfig *config.ApplicationConfig) (schema.TokenizeResponse, error) {
 
-	modelFile := backendConfig.Model
-
 	var inferenceModel grpc.Backend
 	var err error
 
-	opts := ModelOptions(backendConfig, appConfig, model.WithModel(modelFile))
-
+	opts := ModelOptions(backendConfig, appConfig)
 	inferenceModel, err = loader.Load(opts...)
+
 	if err != nil {
 		return schema.TokenizeResponse{}, err
 	}

core/backend/transcript.go

@@ -47,7 +47,7 @@ func ModelTranscription(audio, language string, translate bool, ml *model.ModelL
 			tks = append(tks, int(t))
 		}
 		tr.Segments = append(tr.Segments,
-			schema.Segment{
+			schema.TranscriptionSegment{
 				Text:   s.Text,
 				Id:     int(s.Id),
 				Start:  time.Duration(s.Start),

core/backend/tts.go

@@ -14,28 +14,22 @@ import (
 )
 
 func ModelTTS(
-	backend,
 	text,
-	modelFile,
 	voice,
 	language string,
 	loader *model.ModelLoader,
 	appConfig *config.ApplicationConfig,
 	backendConfig config.BackendConfig,
 ) (string, *proto.Result, error) {
-	bb := backend
-	if bb == "" {
-		bb = model.PiperBackend
-	}
-
-	opts := ModelOptions(backendConfig, appConfig, model.WithBackendString(bb), model.WithModel(modelFile))
+	opts := ModelOptions(backendConfig, appConfig, model.WithDefaultBackendString(model.PiperBackend))
 	ttsModel, err := loader.Load(opts...)
+
 	if err != nil {
 		return "", nil, err
 	}
 
 	if ttsModel == nil {
-		return "", nil, fmt.Errorf("could not load piper model")
+		return "", nil, fmt.Errorf("could not load tts model %q", backendConfig.Model)
 	}
 
 	if err := os.MkdirAll(appConfig.AudioDir, 0750); err != nil {
@@ -45,22 +39,21 @@ func ModelTTS(
 	fileName := utils.GenerateUniqueFileName(appConfig.AudioDir, "tts", ".wav")
 	filePath := filepath.Join(appConfig.AudioDir, fileName)
 
-	// If the model file is not empty, we pass it joined with the model path
+	// We join the model name to the model path here. This seems to only be done for TTS and is HIGHLY suspect.
+	// This should be addressed in a follow up PR soon.
+	// Copying it over nearly verbatim, as TTS backends are not functional without this.
 	modelPath := ""
-	if modelFile != "" {
-		// If the model file is not empty, we pass it joined with the model path
-		// Checking first that it exists and is not outside ModelPath
-		// TODO: we should actually first check if the modelFile is looking like
-		// a FS path
-		mp := filepath.Join(loader.ModelPath, modelFile)
-		if _, err := os.Stat(mp); err == nil {
-			if err := utils.VerifyPath(mp, appConfig.ModelPath); err != nil {
-				return "", nil, err
-			}
-			modelPath = mp
-		} else {
-			modelPath = modelFile
+	// Checking first that it exists and is not outside ModelPath
+	// TODO: we should actually first check if the modelFile is looking like
+	// a FS path
+	mp := filepath.Join(loader.ModelPath, backendConfig.Model)
+	if _, err := os.Stat(mp); err == nil {
+		if err := utils.VerifyPath(mp, appConfig.ModelPath); err != nil {
+			return "", nil, err
 		}
+		modelPath = mp
+	} else {
+		modelPath = backendConfig.Model // skip this step if it fails?????
 	}
 
 	res, err := ttsModel.TTS(context.Background(), &proto.TTSRequest{

core/backend/vad.go

@@ -0,0 +1,38 @@
+package backend
+
+import (
+	"context"
+
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/schema"
+	"github.com/mudler/LocalAI/pkg/grpc/proto"
+	"github.com/mudler/LocalAI/pkg/model"
+)
+
+func VAD(request *schema.VADRequest,
+	ctx context.Context,
+	ml *model.ModelLoader,
+	appConfig *config.ApplicationConfig,
+	backendConfig config.BackendConfig) (*schema.VADResponse, error) {
+	opts := ModelOptions(backendConfig, appConfig)
+	vadModel, err := ml.Load(opts...)
+	if err != nil {
+		return nil, err
+	}
+	req := proto.VADRequest{
+		Audio: request.Audio,
+	}
+	resp, err := vadModel.VAD(ctx, &req)
+	if err != nil {
+		return nil, err
+	}
+
+	segments := []schema.VADSegment{}
+	for _, s := range resp.Segments {
+		segments = append(segments, schema.VADSegment{Start: s.Start, End: s.End})
+	}
+
+	return &schema.VADResponse{
+		Segments: segments,
+	}, nil
+}

core/cli/soundgeneration.go

@@ -86,13 +86,14 @@ func (t *SoundGenerationCMD) Run(ctx *cliContext.Context) error {
 	options := config.BackendConfig{}
 	options.SetDefaults()
 	options.Backend = t.Backend
+	options.Model = t.Model
 
 	var inputFile *string
 	if t.InputFile != "" {
 		inputFile = &t.InputFile
 	}
 
-	filePath, _, err := backend.SoundGeneration(t.Model, text,
+	filePath, _, err := backend.SoundGeneration(text,
 		parseToFloat32Ptr(t.Duration), parseToFloat32Ptr(t.Temperature), &t.DoSample,
 		inputFile, parseToInt32Ptr(t.InputFileSampleDivisor), ml, opts, options)
 

core/cli/tts.go

@@ -52,8 +52,10 @@ func (t *TTSCMD) Run(ctx *cliContext.Context) error {
 
 	options := config.BackendConfig{}
 	options.SetDefaults()
+	options.Backend = t.Backend
+	options.Model = t.Model
 
-	filePath, _, err := backend.ModelTTS(t.Backend, text, t.Model, t.Voice, t.Language, ml, opts, options)
+	filePath, _, err := backend.ModelTTS(text, t.Voice, t.Language, ml, opts, options)
 	if err != nil {
 		return err
 	}

core/config/backend_config.go

@@ -130,25 +130,28 @@ type LLMConfig struct {
 	TrimSpace       []string `yaml:"trimspace"`
 	TrimSuffix      []string `yaml:"trimsuffix"`
 
-	ContextSize          *int      `yaml:"context_size"`
-	NUMA                 bool      `yaml:"numa"`
-	LoraAdapter          string    `yaml:"lora_adapter"`
-	LoraBase             string    `yaml:"lora_base"`
-	LoraAdapters         []string  `yaml:"lora_adapters"`
-	LoraScales           []float32 `yaml:"lora_scales"`
-	LoraScale            float32   `yaml:"lora_scale"`
-	NoMulMatQ            bool      `yaml:"no_mulmatq"`
-	DraftModel           string    `yaml:"draft_model"`
-	NDraft               int32     `yaml:"n_draft"`
-	Quantization         string    `yaml:"quantization"`
-	LoadFormat           string    `yaml:"load_format"`
-	GPUMemoryUtilization float32   `yaml:"gpu_memory_utilization"` // vLLM
-	TrustRemoteCode      bool      `yaml:"trust_remote_code"`      // vLLM
-	EnforceEager         bool      `yaml:"enforce_eager"`          // vLLM
-	SwapSpace            int       `yaml:"swap_space"`             // vLLM
-	MaxModelLen          int       `yaml:"max_model_len"`          // vLLM
-	TensorParallelSize   int       `yaml:"tensor_parallel_size"`   // vLLM
-	MMProj               string    `yaml:"mmproj"`
+	ContextSize          *int             `yaml:"context_size"`
+	NUMA                 bool             `yaml:"numa"`
+	LoraAdapter          string           `yaml:"lora_adapter"`
+	LoraBase             string           `yaml:"lora_base"`
+	LoraAdapters         []string         `yaml:"lora_adapters"`
+	LoraScales           []float32        `yaml:"lora_scales"`
+	LoraScale            float32          `yaml:"lora_scale"`
+	NoMulMatQ            bool             `yaml:"no_mulmatq"`
+	DraftModel           string           `yaml:"draft_model"`
+	NDraft               int32            `yaml:"n_draft"`
+	Quantization         string           `yaml:"quantization"`
+	LoadFormat           string           `yaml:"load_format"`
+	GPUMemoryUtilization float32          `yaml:"gpu_memory_utilization"` // vLLM
+	TrustRemoteCode      bool             `yaml:"trust_remote_code"`      // vLLM
+	EnforceEager         bool             `yaml:"enforce_eager"`          // vLLM
+	SwapSpace            int              `yaml:"swap_space"`             // vLLM
+	MaxModelLen          int              `yaml:"max_model_len"`          // vLLM
+	TensorParallelSize   int              `yaml:"tensor_parallel_size"`   // vLLM
+	DisableLogStatus     bool             `yaml:"disable_log_stats"`      // vLLM
+	DType                string           `yaml:"dtype"`                  // vLLM
+	LimitMMPerPrompt     LimitMMPerPrompt `yaml:"limit_mm_per_prompt"`    // vLLM
+	MMProj               string           `yaml:"mmproj"`
 
 	FlashAttention bool   `yaml:"flash_attention"`
 	NoKVOffloading bool   `yaml:"no_kv_offloading"`
@@ -166,6 +169,13 @@ type LLMConfig struct {
 	CFGScale float32 `yaml:"cfg_scale"` // Classifier-Free Guidance Scale
 }
 
+// LimitMMPerPrompt is a struct that holds the configuration for the limit-mm-per-prompt config in vLLM
+type LimitMMPerPrompt struct {
+	LimitImagePerPrompt int `yaml:"image"`
+	LimitVideoPerPrompt int `yaml:"video"`
+	LimitAudioPerPrompt int `yaml:"audio"`
+}
+
 // AutoGPTQ is a struct that holds the configuration specific to the AutoGPTQ backend
 type AutoGPTQ struct {
 	ModelBaseName    string `yaml:"model_base_name"`
@@ -203,6 +213,8 @@ type TemplateConfig struct {
 	Multimodal string `yaml:"multimodal"`
 
 	JinjaTemplate bool `yaml:"jinja_template"`
+
+	ReplyPrefix string `yaml:"reply_prefix"`
 }
 
 func (c *BackendConfig) UnmarshalYAML(value *yaml.Node) error {
@@ -212,7 +224,15 @@ func (c *BackendConfig) UnmarshalYAML(value *yaml.Node) error {
 		return err
 	}
 	*c = BackendConfig(aux)
+
 	c.KnownUsecases = GetUsecasesFromYAML(c.KnownUsecaseStrings)
+	// Make sure the usecases are valid, we rewrite with what we identified
+	c.KnownUsecaseStrings = []string{}
+	for k, usecase := range GetAllBackendConfigUsecases() {
+		if c.HasUsecases(usecase) {
+			c.KnownUsecaseStrings = append(c.KnownUsecaseStrings, k)
+		}
+	}
 	return nil
 }
 
@@ -437,19 +457,21 @@ func (c *BackendConfig) HasTemplate() bool {
 type BackendConfigUsecases int
 
 const (
-	FLAG_ANY              BackendConfigUsecases = 0b000000000
-	FLAG_CHAT             BackendConfigUsecases = 0b000000001
-	FLAG_COMPLETION       BackendConfigUsecases = 0b000000010
-	FLAG_EDIT             BackendConfigUsecases = 0b000000100
-	FLAG_EMBEDDINGS       BackendConfigUsecases = 0b000001000
-	FLAG_RERANK           BackendConfigUsecases = 0b000010000
-	FLAG_IMAGE            BackendConfigUsecases = 0b000100000
-	FLAG_TRANSCRIPT       BackendConfigUsecases = 0b001000000
-	FLAG_TTS              BackendConfigUsecases = 0b010000000
-	FLAG_SOUND_GENERATION BackendConfigUsecases = 0b100000000
+	FLAG_ANY              BackendConfigUsecases = 0b00000000000
+	FLAG_CHAT             BackendConfigUsecases = 0b00000000001
+	FLAG_COMPLETION       BackendConfigUsecases = 0b00000000010
+	FLAG_EDIT             BackendConfigUsecases = 0b00000000100
+	FLAG_EMBEDDINGS       BackendConfigUsecases = 0b00000001000
+	FLAG_RERANK           BackendConfigUsecases = 0b00000010000
+	FLAG_IMAGE            BackendConfigUsecases = 0b00000100000
+	FLAG_TRANSCRIPT       BackendConfigUsecases = 0b00001000000
+	FLAG_TTS              BackendConfigUsecases = 0b00010000000
+	FLAG_SOUND_GENERATION BackendConfigUsecases = 0b00100000000
+	FLAG_TOKENIZE         BackendConfigUsecases = 0b01000000000
+	FLAG_VAD              BackendConfigUsecases = 0b10000000000
 
 	// Common Subsets
-	FLAG_LLM BackendConfigUsecases = FLAG_CHAT & FLAG_COMPLETION & FLAG_EDIT
+	FLAG_LLM BackendConfigUsecases = FLAG_CHAT | FLAG_COMPLETION | FLAG_EDIT
 )
 
 func GetAllBackendConfigUsecases() map[string]BackendConfigUsecases {
@@ -464,10 +486,16 @@ func GetAllBackendConfigUsecases() map[string]BackendConfigUsecases {
 		"FLAG_TRANSCRIPT":       FLAG_TRANSCRIPT,
 		"FLAG_TTS":              FLAG_TTS,
 		"FLAG_SOUND_GENERATION": FLAG_SOUND_GENERATION,
+		"FLAG_TOKENIZE":         FLAG_TOKENIZE,
+		"FLAG_VAD":              FLAG_VAD,
 		"FLAG_LLM":              FLAG_LLM,
 	}
 }
 
+func stringToFlag(s string) string {
+	return "FLAG_" + strings.ToUpper(s)
+}
+
 func GetUsecasesFromYAML(input []string) *BackendConfigUsecases {
 	if len(input) == 0 {
 		return nil
@@ -475,7 +503,7 @@ func GetUsecasesFromYAML(input []string) *BackendConfigUsecases {
 	result := FLAG_ANY
 	flags := GetAllBackendConfigUsecases()
 	for _, str := range input {
-		flag, exists := flags["FLAG_"+strings.ToUpper(str)]
+		flag, exists := flags[stringToFlag(str)]
 		if exists {
 			result |= flag
 		}
@@ -549,5 +577,18 @@ func (c *BackendConfig) GuessUsecases(u BackendConfigUsecases) bool {
 		}
 	}
 
+	if (u & FLAG_TOKENIZE) == FLAG_TOKENIZE {
+		tokenizeCapableBackends := []string{"llama.cpp", "rwkv"}
+		if !slices.Contains(tokenizeCapableBackends, c.Backend) {
+			return false
+		}
+	}
+
+	if (u & FLAG_VAD) == FLAG_VAD {
+		if c.Backend != "silero-vad" {
+			return false
+		}
+	}
+
 	return true
 }

core/config/backend_config_loader.go

@@ -81,10 +81,10 @@ func readMultipleBackendConfigsFromFile(file string, opts ...ConfigLoaderOption)
 	c := &[]*BackendConfig{}
 	f, err := os.ReadFile(file)
 	if err != nil {
-		return nil, fmt.Errorf("cannot read config file: %w", err)
+		return nil, fmt.Errorf("readMultipleBackendConfigsFromFile cannot read config file %q: %w", file, err)
 	}
 	if err := yaml.Unmarshal(f, c); err != nil {
-		return nil, fmt.Errorf("cannot unmarshal config file: %w", err)
+		return nil, fmt.Errorf("readMultipleBackendConfigsFromFile cannot unmarshal config file %q: %w", file, err)
 	}
 
 	for _, cc := range *c {
@@ -101,10 +101,10 @@ func readBackendConfigFromFile(file string, opts ...ConfigLoaderOption) (*Backen
 	c := &BackendConfig{}
 	f, err := os.ReadFile(file)
 	if err != nil {
-		return nil, fmt.Errorf("cannot read config file: %w", err)
+		return nil, fmt.Errorf("readBackendConfigFromFile cannot read config file %q: %w", file, err)
 	}
 	if err := yaml.Unmarshal(f, c); err != nil {
-		return nil, fmt.Errorf("cannot unmarshal config file: %w", err)
+		return nil, fmt.Errorf("readBackendConfigFromFile cannot unmarshal config file %q: %w", file, err)
 	}
 
 	c.SetDefaults(opts...)
@@ -117,7 +117,9 @@ func (bcl *BackendConfigLoader) LoadBackendConfigFileByName(modelName, modelPath
 	// Load a config file if present after the model name
 	cfg := &BackendConfig{
 		PredictionOptions: schema.PredictionOptions{
-			Model: modelName,
+			BasicModelRequest: schema.BasicModelRequest{
+				Model: modelName,
+			},
 		},
 	}
 
@@ -145,6 +147,15 @@ func (bcl *BackendConfigLoader) LoadBackendConfigFileByName(modelName, modelPath
 	return cfg, nil
 }
 
+func (bcl *BackendConfigLoader) LoadBackendConfigFileByNameDefaultOptions(modelName string, appConfig *ApplicationConfig) (*BackendConfig, error) {
+	return bcl.LoadBackendConfigFileByName(modelName, appConfig.ModelPath,
+		LoadOptionDebug(appConfig.Debug),
+		LoadOptionThreads(appConfig.Threads),
+		LoadOptionContextSize(appConfig.ContextSize),
+		LoadOptionF16(appConfig.F16),
+		ModelPath(appConfig.ModelPath))
+}
+
 // This format is currently only used when reading a single file at startup, passed in via ApplicationConfig.ConfigFile
 func (bcl *BackendConfigLoader) LoadMultipleBackendConfigsSingleFile(file string, opts ...ConfigLoaderOption) error {
 	bcl.Lock()
@@ -167,7 +178,7 @@ func (bcl *BackendConfigLoader) LoadBackendConfig(file string, opts ...ConfigLoa
 	defer bcl.Unlock()
 	c, err := readBackendConfigFromFile(file, opts...)
 	if err != nil {
-		return fmt.Errorf("cannot read config file: %w", err)
+		return fmt.Errorf("LoadBackendConfig cannot read config file %q: %w", file, err)
 	}
 
 	if c.Validate() {
@@ -324,9 +335,10 @@ func (bcl *BackendConfigLoader) Preload(modelPath string) error {
 func (bcl *BackendConfigLoader) LoadBackendConfigsFromPath(path string, opts ...ConfigLoaderOption) error {
 	bcl.Lock()
 	defer bcl.Unlock()
+
 	entries, err := os.ReadDir(path)
 	if err != nil {
-		return fmt.Errorf("cannot read directory '%s': %w", path, err)
+		return fmt.Errorf("LoadBackendConfigsFromPath cannot read directory '%s': %w", path, err)
 	}
 	files := make([]fs.FileInfo, 0, len(entries))
 	for _, entry := range entries {
@@ -344,13 +356,13 @@ func (bcl *BackendConfigLoader) LoadBackendConfigsFromPath(path string, opts ...
 		}
 		c, err := readBackendConfigFromFile(filepath.Join(path, file.Name()), opts...)
 		if err != nil {
-			log.Error().Err(err).Msgf("cannot read config file: %s", file.Name())
+			log.Error().Err(err).Str("File Name", file.Name()).Msgf("LoadBackendConfigsFromPath cannot read config file")
 			continue
 		}
 		if c.Validate() {
 			bcl.configs[c.Name] = *c
 		} else {
-			log.Error().Err(err).Msgf("config is not valid")
+			log.Error().Err(err).Str("Name", c.Name).Msgf("config is not valid")
 		}
 	}
 

core/config/guesser.go

@@ -161,10 +161,11 @@ func guessDefaultsFromFile(cfg *BackendConfig, modelPath string) {
 	}
 
 	// We try to guess only if we don't have a template defined already
-	f, err := gguf.ParseGGUFFile(filepath.Join(modelPath, cfg.ModelFileName()))
+	guessPath := filepath.Join(modelPath, cfg.ModelFileName())
+	f, err := gguf.ParseGGUFFile(guessPath)
 	if err != nil {
 		// Only valid for gguf files
-		log.Debug().Msgf("guessDefaultsFromFile: %s", "not a GGUF file")
+		log.Debug().Str("filePath", guessPath).Msg("guessDefaultsFromFile: not a GGUF file")
 		return
 	}
 

core/gallery/gallery.go

@@ -29,6 +29,8 @@ func InstallModelFromGallery(galleries []config.Gallery, name string, basePath s
 			if err != nil {
 				return err
 			}
+			config.Description = model.Description
+			config.License = model.License
 		} else if len(model.ConfigFile) > 0 {
 			// TODO: is this worse than using the override method with a blank cfg yaml?
 			reYamlConfig, err := yaml.Marshal(model.ConfigFile)
@@ -114,7 +116,7 @@ func FindModel(models []*GalleryModel, name string, basePath string) *GalleryMod
 // List available models
 // Models galleries are a list of yaml files that are hosted on a remote server (for example github).
 // Each yaml file contains a list of models that can be downloaded and optionally overrides to define a new model setting.
-func AvailableGalleryModels(galleries []config.Gallery, basePath string) ([]*GalleryModel, error) {
+func AvailableGalleryModels(galleries []config.Gallery, basePath string) (GalleryModels, error) {
 	var models []*GalleryModel
 
 	// Get models from galleries

core/gallery/request.go

@@ -62,3 +62,15 @@ func (gm GalleryModels) FindByName(name string) *GalleryModel {
 	}
 	return nil
 }
+
+func (gm GalleryModels) Paginate(pageNum int, itemsNum int) GalleryModels {
+	start := (pageNum - 1) * itemsNum
+	end := start + itemsNum
+	if start > len(gm) {
+		start = len(gm)
+	}
+	if end > len(gm) {
+		end = len(gm)
+	}
+	return gm[start:end]
+}

core/http/app.go

@@ -130,7 +130,6 @@ func API(application *application.Application) (*fiber.App, error) {
 				return metricsService.Shutdown()
 			})
 		}
-
 	}
 	// Health Checks should always be exempt from auth, so register these first
 	routes.HealthRoutes(router)
@@ -140,6 +139,28 @@ func API(application *application.Application) (*fiber.App, error) {
 		return nil, fmt.Errorf("failed to create key auth config: %w", err)
 	}
 
+	httpFS := http.FS(embedDirStatic)
+
+	router.Use(favicon.New(favicon.Config{
+		URL:        "/favicon.ico",
+		FileSystem: httpFS,
+		File:       "static/favicon.ico",
+	}))
+
+	router.Use("/static", filesystem.New(filesystem.Config{
+		Root:       httpFS,
+		PathPrefix: "static",
+		Browse:     true,
+	}))
+
+	if application.ApplicationConfig().ImageDir != "" {
+		router.Static("/generated-images", application.ApplicationConfig().ImageDir)
+	}
+
+	if application.ApplicationConfig().AudioDir != "" {
+		router.Static("/generated-audio", application.ApplicationConfig().AudioDir)
+	}
+
 	// Auth is applied to _all_ endpoints. No exceptions. Filtering out endpoints to bypass is the role of the Filter property of the KeyAuth Configuration
 	router.Use(v2keyauth.New(*kaConfig))
 
@@ -167,27 +188,15 @@ func API(application *application.Application) (*fiber.App, error) {
 	galleryService := services.NewGalleryService(application.ApplicationConfig())
 	galleryService.Start(application.ApplicationConfig().Context, application.BackendLoader())
 
-	routes.RegisterElevenLabsRoutes(router, application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig())
-	routes.RegisterLocalAIRoutes(router, application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig(), galleryService)
-	routes.RegisterOpenAIRoutes(router, application)
+	requestExtractor := middleware.NewRequestExtractor(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig())
+
+	routes.RegisterElevenLabsRoutes(router, requestExtractor, application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig())
+	routes.RegisterLocalAIRoutes(router, requestExtractor, application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig(), galleryService)
+	routes.RegisterOpenAIRoutes(router, requestExtractor, application)
 	if !application.ApplicationConfig().DisableWebUI {
 		routes.RegisterUIRoutes(router, application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig(), galleryService)
 	}
-	routes.RegisterJINARoutes(router, application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig())
-
-	httpFS := http.FS(embedDirStatic)
-
-	router.Use(favicon.New(favicon.Config{
-		URL:        "/favicon.ico",
-		FileSystem: httpFS,
-		File:       "static/favicon.ico",
-	}))
-
-	router.Use("/static", filesystem.New(filesystem.Config{
-		Root:       httpFS,
-		PathPrefix: "static",
-		Browse:     true,
-	}))
+	routes.RegisterJINARoutes(router, requestExtractor, application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig())
 
 	// Define a custom 404 handler
 	// Note: keep this at the bottom!

core/http/ctx/fiber.go

@@ -1,47 +0,0 @@
-package fiberContext
-
-import (
-	"fmt"
-	"strings"
-
-	"github.com/gofiber/fiber/v2"
-	"github.com/mudler/LocalAI/core/config"
-	"github.com/mudler/LocalAI/core/services"
-	"github.com/mudler/LocalAI/pkg/model"
-	"github.com/rs/zerolog/log"
-)
-
-// ModelFromContext returns the model from the context
-// If no model is specified, it will take the first available
-// Takes a model string as input which should be the one received from the user request.
-// It returns the model name resolved from the context and an error if any.
-func ModelFromContext(ctx *fiber.Ctx, cl *config.BackendConfigLoader, loader *model.ModelLoader, modelInput string, firstModel bool) (string, error) {
-	if ctx.Params("model") != "" {
-		modelInput = ctx.Params("model")
-	}
-	if ctx.Query("model") != "" {
-		modelInput = ctx.Query("model")
-	}
-	// Set model from bearer token, if available
-	bearer := strings.TrimLeft(ctx.Get("authorization"), "Bear ") // Reduced duplicate characters of Bearer
-	bearerExists := bearer != "" && loader.ExistsInModelPath(bearer)
-
-	// If no model was specified, take the first available
-	if modelInput == "" && !bearerExists && firstModel {
-		models, _ := services.ListModels(cl, loader, config.NoFilterFn, services.SKIP_IF_CONFIGURED)
-		if len(models) > 0 {
-			modelInput = models[0]
-			log.Debug().Msgf("No model specified, using: %s", modelInput)
-		} else {
-			log.Debug().Msgf("No model specified, returning error")
-			return "", fmt.Errorf("no model specified")
-		}
-	}
-
-	// If a model is found in bearer token takes precedence
-	if bearerExists {
-		log.Debug().Msgf("Using model from bearer token: %s", bearer)
-		modelInput = bearer
-	}
-	return modelInput, nil
-}

core/http/elements/buttons.go

@@ -13,7 +13,7 @@ func installButton(galleryName string) elem.Node {
 		attrs.Props{
 			"data-twe-ripple-init":  "",
 			"data-twe-ripple-color": "light",
-			"class":                 "float-right inline-block rounded bg-primary px-6 pb-2.5 mb-3 pt-2.5 text-xs font-medium uppercase leading-normal text-white shadow-primary-3 transition duration-150 ease-in-out hover:bg-primary-accent-300 hover:shadow-primary-2 focus:bg-primary-accent-300 focus:shadow-primary-2 focus:outline-none focus:ring-0 active:bg-primary-600 active:shadow-primary-2 dark:shadow-black/30 dark:hover:shadow-dark-strong dark:focus:shadow-dark-strong dark:active:shadow-dark-strong",
+			"class":                 "float-right inline-flex items-center rounded-lg bg-blue-600 hover:bg-blue-700 px-4 py-2 text-sm font-medium text-white transition duration-300 ease-in-out shadow hover:shadow-lg",
 			"hx-swap":               "outerHTML",
 			// post the Model ID as param
 			"hx-post": "browse/install/model/" + galleryName,
@@ -52,7 +52,7 @@ func infoButton(m *gallery.GalleryModel) elem.Node {
 		attrs.Props{
 			"data-twe-ripple-init":  "",
 			"data-twe-ripple-color": "light",
-			"class":                 "float-left inline-block rounded bg-primary px-6 pb-2.5 mb-3 pt-2.5 text-xs font-medium uppercase leading-normal text-white shadow-primary-3 transition duration-150 ease-in-out hover:bg-primary-accent-300 hover:shadow-primary-2 focus:bg-primary-accent-300 focus:shadow-primary-2 focus:outline-none focus:ring-0 active:bg-primary-600 active:shadow-primary-2 dark:shadow-black/30 dark:hover:shadow-dark-strong dark:focus:shadow-dark-strong dark:active:shadow-dark-strong",
+			"class":                 "inline-flex items-center rounded-lg bg-gray-700 hover:bg-gray-600 px-4 py-2 text-sm font-medium text-white transition duration-300 ease-in-out",
 			"data-modal-target":     modalName(m),
 			"data-modal-toggle":     modalName(m),
 		},

core/http/elements/gallery.go

@@ -17,7 +17,7 @@ const (
 func cardSpan(text, icon string) elem.Node {
 	return elem.Span(
 		attrs.Props{
-			"class": "inline-block bg-gray-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2",
+			"class": "inline-flex items-center px-3 py-1 rounded-lg text-xs font-medium bg-gray-700/70 text-gray-300 border border-gray-600/50 mr-2 mb-2",
 		},
 		elem.I(attrs.Props{
 			"class": icon + " pr-2",
@@ -39,19 +39,20 @@ func searchableElement(text, icon string) elem.Node {
 		),
 		elem.Span(
 			attrs.Props{
-				"class": "inline-block bg-gray-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2 hover:bg-gray-300 hover:shadow-gray-2",
+				"class": "inline-flex items-center text-xs px-3 py-1 rounded-full bg-gray-700/60 text-gray-300 border border-gray-600/50 hover:bg-gray-600 hover:text-gray-100 transition duration-200 ease-in-out",
 			},
 			elem.A(
 				attrs.Props{
 					//	"name":      "search",
 					//	"value":     text,
 					//"class":     "inline-block bg-gray-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2",
-					"href":      "#!",
-					"hx-post":   "browse/search/models",
-					"hx-target": "#search-results",
+					//"href":      "#!",
+					"href": "browse?term=" + text,
+					//"hx-post":   "browse/search/models",
+					//"hx-target": "#search-results",
 					// TODO: this doesn't work
 					//	"hx-vals":      `{ \"search\": \"` + text + `\" }`,
-					"hx-indicator": ".htmx-indicator",
+					//"hx-indicator": ".htmx-indicator",
 				},
 				elem.I(attrs.Props{
 					"class": icon + " pr-2",
@@ -101,7 +102,7 @@ func modalName(m *gallery.GalleryModel) string {
 	return m.Name + "-modal"
 }
 
-func modelDescription(m *gallery.GalleryModel) elem.Node {
+func modelModal(m *gallery.GalleryModel) elem.Node {
 	urls := []elem.Node{}
 	for _, url := range m.URLs {
 		urls = append(urls,
@@ -116,6 +117,125 @@ func modelDescription(m *gallery.GalleryModel) elem.Node {
 		)
 	}
 
+	return elem.Div(
+		attrs.Props{
+			"id":          modalName(m),
+			"tabindex":    "-1",
+			"aria-hidden": "true",
+			"class":       "hidden overflow-y-auto overflow-x-hidden fixed top-0 right-0 left-0 z-50 justify-center items-center w-full md:inset-0 h-[calc(100%-1rem)] max-h-full",
+		},
+		elem.Div(
+			attrs.Props{
+				"class": "relative p-4 w-full max-w-2xl max-h-full",
+			},
+			elem.Div(
+				attrs.Props{
+					"class": "relative p-4 w-full max-w-2xl max-h-full bg-white rounded-lg shadow dark:bg-gray-700",
+				},
+				// header
+				elem.Div(
+					attrs.Props{
+						"class": "flex items-center justify-between p-4 md:p-5 border-b rounded-t dark:border-gray-600",
+					},
+					elem.H3(
+						attrs.Props{
+							"class": "text-xl font-semibold text-gray-900 dark:text-white",
+						},
+						elem.Text(bluemonday.StrictPolicy().Sanitize(m.Name)),
+					),
+					elem.Button( // close button
+						attrs.Props{
+							"class":           "text-gray-400 bg-transparent hover:bg-gray-200 hover:text-gray-900 rounded-lg text-sm w-8 h-8 ms-auto inline-flex justify-center items-center dark:hover:bg-gray-600 dark:hover:text-white",
+							"data-modal-hide": modalName(m),
+						},
+						elem.Raw(
+							`<svg class="w-3 h-3" aria-hidden="true" xmlns="http://www.w3.org/2000/svg" fill="none" viewBox="0 0 14 14">
+							<path stroke="currentColor" stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="m1 1 6 6m0 0 6 6M7 7l6-6M7 7l-6 6"/>
+						</svg>`,
+						),
+						elem.Span(
+							attrs.Props{
+								"class": "sr-only",
+							},
+							elem.Text("Close modal"),
+						),
+					),
+				),
+				// body
+				elem.Div(
+					attrs.Props{
+						"class": "p-4 md:p-5 space-y-4",
+					},
+					elem.Div(
+						attrs.Props{
+							"class": "flex justify-center items-center",
+						},
+						elem.Img(attrs.Props{
+							//	"class": "rounded-t-lg object-fit object-center h-96",
+							"class":   "lazy rounded-t-lg max-h-48 max-w-96 object-cover mt-3 entered loaded",
+							"src":     m.Icon,
+							"loading": "lazy",
+						}),
+					),
+					elem.P(
+						attrs.Props{
+							"class": "text-base leading-relaxed text-gray-500 dark:text-gray-400",
+						},
+						elem.Text(bluemonday.StrictPolicy().Sanitize(m.Description)),
+					),
+					elem.Hr(
+						attrs.Props{},
+					),
+					elem.P(
+						attrs.Props{
+							"class": "text-sm font-semibold text-gray-900 dark:text-white",
+						},
+						elem.Text("Links"),
+					),
+					elem.Ul(
+						attrs.Props{},
+						urls...,
+					),
+					elem.If(
+						len(m.Tags) > 0,
+						elem.Div(
+							attrs.Props{},
+							elem.P(
+								attrs.Props{
+									"class": "text-sm mb-5 font-semibold text-gray-900 dark:text-white",
+								},
+								elem.Text("Tags"),
+							),
+							elem.Div(
+								attrs.Props{
+									"class": "flex flex-row flex-wrap content-center",
+								},
+								tagsNodes...,
+							),
+						),
+						elem.Div(attrs.Props{}),
+					),
+				),
+				// Footer
+				elem.Div(
+					attrs.Props{
+						"class": "flex items-center p-4 md:p-5 border-t border-gray-200 rounded-b dark:border-gray-600",
+					},
+					elem.Button(
+						attrs.Props{
+							"data-modal-hide": modalName(m),
+							"class":           "py-2.5 px-5 ms-3 text-sm font-medium text-gray-900 focus:outline-none bg-white rounded-lg border border-gray-200 hover:bg-gray-100 hover:text-blue-700 focus:z-10 focus:ring-4 focus:ring-gray-100 dark:focus:ring-gray-700 dark:bg-gray-800 dark:text-gray-400 dark:border-gray-600 dark:hover:text-white dark:hover:bg-gray-700",
+						},
+						elem.Text("Close"),
+					),
+				),
+			),
+		),
+	)
+
+}
+
+func modelDescription(m *gallery.GalleryModel) elem.Node {
 	return elem.Div(
 		attrs.Props{
 			"class": "p-6 text-surface dark:text-white",
@@ -132,122 +252,6 @@ func modelDescription(m *gallery.GalleryModel) elem.Node {
 			},
 			elem.Text(bluemonday.StrictPolicy().Sanitize(m.Description)),
 		),
-
-		elem.Div(
-			attrs.Props{
-				"id":          modalName(m),
-				"tabindex":    "-1",
-				"aria-hidden": "true",
-				"class":       "hidden overflow-y-auto overflow-x-hidden fixed top-0 right-0 left-0 z-50 justify-center items-center w-full md:inset-0 h-[calc(100%-1rem)] max-h-full",
-			},
-			elem.Div(
-				attrs.Props{
-					"class": "relative p-4 w-full max-w-2xl max-h-full",
-				},
-				elem.Div(
-					attrs.Props{
-						"class": "relative p-4 w-full max-w-2xl max-h-full bg-white rounded-lg shadow dark:bg-gray-700",
-					},
-					// header
-					elem.Div(
-						attrs.Props{
-							"class": "flex items-center justify-between p-4 md:p-5 border-b rounded-t dark:border-gray-600",
-						},
-						elem.H3(
-							attrs.Props{
-								"class": "text-xl font-semibold text-gray-900 dark:text-white",
-							},
-							elem.Text(bluemonday.StrictPolicy().Sanitize(m.Name)),
-						),
-						elem.Button( // close button
-							attrs.Props{
-								"class":           "text-gray-400 bg-transparent hover:bg-gray-200 hover:text-gray-900 rounded-lg text-sm w-8 h-8 ms-auto inline-flex justify-center items-center dark:hover:bg-gray-600 dark:hover:text-white",
-								"data-modal-hide": modalName(m),
-							},
-							elem.Raw(
-								`<svg class="w-3 h-3" aria-hidden="true" xmlns="http://www.w3.org/2000/svg" fill="none" viewBox="0 0 14 14">
-									<path stroke="currentColor" stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="m1 1 6 6m0 0 6 6M7 7l6-6M7 7l-6 6"/>
-								</svg>`,
-							),
-							elem.Span(
-								attrs.Props{
-									"class": "sr-only",
-								},
-								elem.Text("Close modal"),
-							),
-						),
-					),
-					// body
-					elem.Div(
-						attrs.Props{
-							"class": "p-4 md:p-5 space-y-4",
-						},
-						elem.Div(
-							attrs.Props{
-								"class": "flex justify-center items-center",
-							},
-							elem.Img(attrs.Props{
-								//	"class": "rounded-t-lg object-fit object-center h-96",
-								"class":   "lazy rounded-t-lg max-h-48 max-w-96 object-cover mt-3 entered loaded",
-								"src":     m.Icon,
-								"loading": "lazy",
-							}),
-						),
-						elem.P(
-							attrs.Props{
-								"class": "text-base leading-relaxed text-gray-500 dark:text-gray-400",
-							},
-							elem.Text(bluemonday.StrictPolicy().Sanitize(m.Description)),
-						),
-						elem.Hr(
-							attrs.Props{},
-						),
-						elem.P(
-							attrs.Props{
-								"class": "text-sm font-semibold text-gray-900 dark:text-white",
-							},
-							elem.Text("Links"),
-						),
-						elem.Ul(
-							attrs.Props{},
-							urls...,
-						),
-						elem.If(
-							len(m.Tags) > 0,
-							elem.Div(
-								attrs.Props{},
-								elem.P(
-									attrs.Props{
-										"class": "text-sm mb-5 font-semibold text-gray-900 dark:text-white",
-									},
-									elem.Text("Tags"),
-								),
-								elem.Div(
-									attrs.Props{
-										"class": "flex flex-row flex-wrap content-center",
-									},
-									tagsNodes...,
-								),
-							),
-							elem.Div(attrs.Props{}),
-						),
-					),
-					// Footer
-					elem.Div(
-						attrs.Props{
-							"class": "flex items-center p-4 md:p-5 border-t border-gray-200 rounded-b dark:border-gray-600",
-						},
-						elem.Button(
-							attrs.Props{
-								"data-modal-hide": modalName(m),
-								"class":           "py-2.5 px-5 ms-3 text-sm font-medium text-gray-900 focus:outline-none bg-white rounded-lg border border-gray-200 hover:bg-gray-100 hover:text-blue-700 focus:z-10 focus:ring-4 focus:ring-gray-100 dark:focus:ring-gray-700 dark:bg-gray-800 dark:text-gray-400 dark:border-gray-600 dark:hover:text-white dark:hover:bg-gray-700",
-							},
-							elem.Text("Close"),
-						),
-					),
-				),
-			),
-		),
 	)
 }
 
@@ -397,7 +401,7 @@ func ListModels(models []*gallery.GalleryModel, processTracker ProcessTracker, g
 		modelsElements = append(modelsElements,
 			elem.Div(
 				attrs.Props{
-					"class": " me-4 mb-2 block rounded-lg bg-white shadow-secondary-1  dark:bg-gray-800 dark:bg-surface-dark dark:text-white text-surface pb-2",
+					"class": " me-4 mb-2 block rounded-lg bg-white shadow-secondary-1  dark:bg-gray-800 dark:bg-surface-dark dark:text-white text-surface pb-2 bg-gray-800/90 border border-gray-700/50 rounded-xl overflow-hidden transition-all duration-300 hover:shadow-lg hover:shadow-blue-900/20 hover:-translate-y-1 hover:border-blue-700/50",
 				},
 				elem.Div(
 					attrs.Props{
@@ -406,6 +410,7 @@ func ListModels(models []*gallery.GalleryModel, processTracker ProcessTracker, g
 					elems...,
 				),
 			),
+			modelModal(m),
 		)
 	}
 

core/http/elements/p2p.go

@@ -2,6 +2,7 @@ package elements
 
 import (
 	"fmt"
+	"time"
 
 	"github.com/chasefleming/elem-go"
 	"github.com/chasefleming/elem-go/attrs"
@@ -18,19 +19,6 @@ func renderElements(n []elem.Node) string {
 }
 
 func P2PNodeStats(nodes []p2p.NodeData) string {
-	/*
-	   <div class="bg-gray-800 p-6 rounded-lg shadow-lg text-left">
-	                       <p class="text-xl font-semibold text-gray-200">Total Workers Detected: {{ len .Nodes }}</p>
-	                       {{ $online := 0 }}
-	                       {{ range .Nodes }}
-	                           {{ if .IsOnline }}
-	                               {{ $online = add $online 1 }}
-	                           {{ end }}
-	                       {{ end }}
-	                       <p class="text-xl font-semibold text-gray-200">Total Online Workers: {{$online}}</p>
-	                   </div>
-	*/
-
 	online := 0
 	for _, n := range nodes {
 		if n.IsOnline() {
@@ -38,27 +26,21 @@ func P2PNodeStats(nodes []p2p.NodeData) string {
 		}
 	}
 
-	class := "text-green-500"
+	class := "text-blue-400"
 	if online == 0 {
-		class = "text-red-500"
+		class = "text-red-400"
 	}
-	/*
-	   <i class="fas fa-circle animate-pulse text-green-500 ml-2 mr-1"></i>
-	*/
-	circle := elem.I(attrs.Props{
-		"class": "fas fa-circle animate-pulse " + class + " ml-2 mr-1",
-	})
+
 	nodesElements := []elem.Node{
 		elem.Span(
 			attrs.Props{
-				"class": class,
+				"class": class + " font-bold text-xl",
 			},
-			circle,
 			elem.Text(fmt.Sprintf("%d", online)),
 		),
 		elem.Span(
 			attrs.Props{
-				"class": "text-gray-200",
+				"class": "text-gray-300 text-xl",
 			},
 			elem.Text(fmt.Sprintf("/%d", len(nodes))),
 		),
@@ -68,77 +50,73 @@ func P2PNodeStats(nodes []p2p.NodeData) string {
 }
 
 func P2PNodeBoxes(nodes []p2p.NodeData) string {
-	/*
-			<div class="bg-gray-800 p-4 rounded-lg shadow-lg text-left">
-			<div class="flex items-center mb-2">
-				<i class="fas fa-desktop text-gray-400 mr-2"></i>
-				<span class="text-gray-200 font-semibold">{{.ID}}</span>
-			</div>
-			<p class="text-sm text-gray-400 mt-2 flex items-center">
-				Status:
-				<i class="fas fa-circle {{ if .IsOnline }}text-green-500{{ else }}text-red-500{{ end }} ml-2 mr-1"></i>
-				<span class="{{ if .IsOnline }}text-green-400{{ else }}text-red-400{{ end }}">
-					{{ if .IsOnline }}Online{{ else }}Offline{{ end }}
-				</span>
-			</p>
-		</div>
-	*/
-
 	nodesElements := []elem.Node{}
 
 	for _, n := range nodes {
+		nodeID := bluemonday.StrictPolicy().Sanitize(n.ID)
+
+		// Define status-specific classes
+		statusIconClass := "text-green-400"
+		statusText := "Online"
+		statusTextClass := "text-green-400"
+
+		if !n.IsOnline() {
+			statusIconClass = "text-red-400"
+			statusText = "Offline"
+			statusTextClass = "text-red-400"
+		}
 
 		nodesElements = append(nodesElements,
 			elem.Div(
 				attrs.Props{
-					"class": "bg-gray-700 p-6 rounded-lg shadow-lg text-left",
+					"class": "bg-gray-800/80 border border-gray-700/50 rounded-xl p-4 shadow-lg transition-all duration-300 hover:shadow-blue-900/20 hover:border-blue-700/50",
 				},
-				elem.P(
+				// Node ID and status indicator in top row
+				elem.Div(
 					attrs.Props{
-						"class": "text-sm text-gray-400 mt-2 flex",
+						"class": "flex items-center justify-between mb-3",
 					},
-					elem.I(
+					// Node ID with icon
+					elem.Div(
 						attrs.Props{
-							"class": "fas fa-desktop text-gray-400 mr-2",
+							"class": "flex items-center",
 						},
-					),
-					elem.Text("Name: "),
-					elem.Span(
-						attrs.Props{
-							"class": "text-gray-200 font-semibold ml-2 mr-1",
-						},
-						elem.Text(bluemonday.StrictPolicy().Sanitize(n.ID)),
-					),
-					elem.Text("Status: "),
-					elem.If(
-						n.IsOnline(),
 						elem.I(
 							attrs.Props{
-								"class": "fas fa-circle animate-pulse text-green-500 ml-2 mr-1",
+								"class": "fas fa-server text-blue-400 mr-2",
 							},
 						),
-						elem.I(
-							attrs.Props{
-								"class": "fas fa-circle animate-pulse text-red-500 ml-2 mr-1",
-							},
-						),
-					),
-					elem.If(
-						n.IsOnline(),
-						elem.Span(
-							attrs.Props{
-								"class": "text-green-400",
-							},
-
-							elem.Text("Online"),
-						),
 						elem.Span(
 							attrs.Props{
-								"class": "text-red-400",
+								"class": "text-white font-medium",
 							},
-							elem.Text("Offline"),
+							elem.Text(nodeID),
 						),
 					),
+					// Status indicator
+					elem.Div(
+						attrs.Props{
+							"class": "flex items-center",
+						},
+						elem.I(
+							attrs.Props{
+								"class": "fas fa-circle animate-pulse " + statusIconClass + " mr-1.5",
+							},
+						),
+						elem.Span(
+							attrs.Props{
+								"class": statusTextClass,
+							},
+							elem.Text(statusText),
+						),
+					),
+				),
+				// Bottom section with timestamp
+				elem.Div(
+					attrs.Props{
+						"class": "text-xs text-gray-400 pt-1 border-t border-gray-700/30",
+					},
+					elem.Text("Last updated: "+time.Now().UTC().Format("2006-01-02 15:04:05")),
 				),
 			))
 	}

core/http/endpoints/elevenlabs/soundgeneration.go

@@ -4,7 +4,7 @@ import (
 	"github.com/gofiber/fiber/v2"
 	"github.com/mudler/LocalAI/core/backend"
 	"github.com/mudler/LocalAI/core/config"
-	fiberContext "github.com/mudler/LocalAI/core/http/ctx"
+	"github.com/mudler/LocalAI/core/http/middleware"
 	"github.com/mudler/LocalAI/core/schema"
 	"github.com/mudler/LocalAI/pkg/model"
 	"github.com/rs/zerolog/log"
@@ -17,45 +17,21 @@ import (
 // @Router /v1/sound-generation [post]
 func SoundGenerationEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
-		input := new(schema.ElevenLabsSoundGenerationRequest)
-		// Get input data from the request body
-		if err := c.BodyParser(input); err != nil {
-			return err
+
+		input, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_LOCALAI_REQUEST).(*schema.ElevenLabsSoundGenerationRequest)
+		if !ok || input.ModelID == "" {
+			return fiber.ErrBadRequest
 		}
 
-		modelFile, err := fiberContext.ModelFromContext(c, cl, ml, input.ModelID, false)
-		if err != nil {
-			modelFile = input.ModelID
-			log.Warn().Str("ModelID", input.ModelID).Msg("Model not found in context")
+		cfg, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_MODEL_CONFIG).(*config.BackendConfig)
+		if !ok || cfg == nil {
+			return fiber.ErrBadRequest
 		}
 
-		cfg, err := cl.LoadBackendConfigFileByName(modelFile, appConfig.ModelPath,
-			config.LoadOptionDebug(appConfig.Debug),
-			config.LoadOptionThreads(appConfig.Threads),
-			config.LoadOptionContextSize(appConfig.ContextSize),
-			config.LoadOptionF16(appConfig.F16),
-		)
-		if err != nil {
-			modelFile = input.ModelID
-			log.Warn().Str("Request ModelID", input.ModelID).Err(err).Msg("error during LoadBackendConfigFileByName, using request ModelID")
-		} else {
-			if input.ModelID != "" {
-				modelFile = input.ModelID
-			} else {
-				modelFile = cfg.Model
-			}
-		}
 		log.Debug().Str("modelFile", "modelFile").Str("backend", cfg.Backend).Msg("Sound Generation Request about to be sent to backend")
 
-		if input.Duration != nil {
-			log.Debug().Float32("duration", *input.Duration).Msg("duration set")
-		}
-		if input.Temperature != nil {
-			log.Debug().Float32("temperature", *input.Temperature).Msg("temperature set")
-		}
-
 		// TODO: Support uploading files?
-		filePath, _, err := backend.SoundGeneration(modelFile, input.Text, input.Duration, input.Temperature, input.DoSample, nil, nil, ml, appConfig, *cfg)
+		filePath, _, err := backend.SoundGeneration(input.Text, input.Duration, input.Temperature, input.DoSample, nil, nil, ml, appConfig, *cfg)
 		if err != nil {
 			return err
 		}

core/http/endpoints/elevenlabs/tts.go

@@ -3,7 +3,7 @@ package elevenlabs
 import (
 	"github.com/mudler/LocalAI/core/backend"
 	"github.com/mudler/LocalAI/core/config"
-	fiberContext "github.com/mudler/LocalAI/core/http/ctx"
+	"github.com/mudler/LocalAI/core/http/middleware"
 	"github.com/mudler/LocalAI/pkg/model"
 
 	"github.com/gofiber/fiber/v2"
@@ -20,39 +20,21 @@ import (
 func TTSEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
 
-		input := new(schema.ElevenLabsTTSRequest)
 		voiceID := c.Params("voice-id")
 
-		// Get input data from the request body
-		if err := c.BodyParser(input); err != nil {
-			return err
+		input, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_LOCALAI_REQUEST).(*schema.ElevenLabsTTSRequest)
+		if !ok || input.ModelID == "" {
+			return fiber.ErrBadRequest
 		}
 
-		modelFile, err := fiberContext.ModelFromContext(c, cl, ml, input.ModelID, false)
-		if err != nil {
-			modelFile = input.ModelID
-			log.Warn().Msgf("Model not found in context: %s", input.ModelID)
+		cfg, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_MODEL_CONFIG).(*config.BackendConfig)
+		if !ok || cfg == nil {
+			return fiber.ErrBadRequest
 		}
 
-		cfg, err := cl.LoadBackendConfigFileByName(modelFile, appConfig.ModelPath,
-			config.LoadOptionDebug(appConfig.Debug),
-			config.LoadOptionThreads(appConfig.Threads),
-			config.LoadOptionContextSize(appConfig.ContextSize),
-			config.LoadOptionF16(appConfig.F16),
-		)
-		if err != nil {
-			modelFile = input.ModelID
-			log.Warn().Msgf("Model not found in context: %s", input.ModelID)
-		} else {
-			if input.ModelID != "" {
-				modelFile = input.ModelID
-			} else {
-				modelFile = cfg.Model
-			}
-		}
-		log.Debug().Msgf("Request for model: %s", modelFile)
+		log.Debug().Str("modelName", input.ModelID).Msg("elevenlabs TTS request recieved")
 
-		filePath, _, err := backend.ModelTTS(cfg.Backend, input.Text, modelFile, "", voiceID, ml, appConfig, *cfg)
+		filePath, _, err := backend.ModelTTS(input.Text, voiceID, input.LanguageCode, ml, appConfig, *cfg)
 		if err != nil {
 			return err
 		}

core/http/endpoints/jina/rerank.go

@@ -3,9 +3,9 @@ package jina
 import (
 	"github.com/mudler/LocalAI/core/backend"
 	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/http/middleware"
 
 	"github.com/gofiber/fiber/v2"
-	fiberContext "github.com/mudler/LocalAI/core/http/ctx"
 	"github.com/mudler/LocalAI/core/schema"
 	"github.com/mudler/LocalAI/pkg/grpc/proto"
 	"github.com/mudler/LocalAI/pkg/model"
@@ -19,58 +19,32 @@ import (
 // @Router /v1/rerank [post]
 func JINARerankEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
-		req := new(schema.JINARerankRequest)
-		if err := c.BodyParser(req); err != nil {
-			return c.Status(fiber.StatusBadRequest).JSON(fiber.Map{
-				"error": "Cannot parse JSON",
-			})
+
+		input, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_LOCALAI_REQUEST).(*schema.JINARerankRequest)
+		if !ok || input.Model == "" {
+			return fiber.ErrBadRequest
 		}
 
-		input := new(schema.TTSRequest)
-
-		// Get input data from the request body
-		if err := c.BodyParser(input); err != nil {
-			return err
+		cfg, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_MODEL_CONFIG).(*config.BackendConfig)
+		if !ok || cfg == nil {
+			return fiber.ErrBadRequest
 		}
 
-		modelFile, err := fiberContext.ModelFromContext(c, cl, ml, input.Model, false)
-		if err != nil {
-			modelFile = input.Model
-			log.Warn().Msgf("Model not found in context: %s", input.Model)
-		}
-
-		cfg, err := cl.LoadBackendConfigFileByName(modelFile, appConfig.ModelPath,
-			config.LoadOptionDebug(appConfig.Debug),
-			config.LoadOptionThreads(appConfig.Threads),
-			config.LoadOptionContextSize(appConfig.ContextSize),
-			config.LoadOptionF16(appConfig.F16),
-		)
-		if err != nil {
-			modelFile = input.Model
-			log.Warn().Msgf("Model not found in context: %s", input.Model)
-		} else {
-			modelFile = cfg.Model
-		}
-
-		log.Debug().Msgf("Request for model: %s", modelFile)
-
-		if input.Backend != "" {
-			cfg.Backend = input.Backend
-		}
+		log.Debug().Str("model", input.Model).Msg("JINA Rerank Request recieved")
 
 		request := &proto.RerankRequest{
-			Query:     req.Query,
-			TopN:      int32(req.TopN),
-			Documents: req.Documents,
+			Query:     input.Query,
+			TopN:      int32(input.TopN),
+			Documents: input.Documents,
 		}
 
-		results, err := backend.Rerank(modelFile, request, ml, appConfig, *cfg)
+		results, err := backend.Rerank(request, ml, appConfig, *cfg)
 		if err != nil {
 			return err
 		}
 
 		response := &schema.JINARerankResponse{
-			Model: req.Model,
+			Model: input.Model,
 		}
 
 		for _, r := range results.Results {

core/http/endpoints/localai/get_token_metrics.go

@@ -4,13 +4,15 @@ import (
 	"github.com/gofiber/fiber/v2"
 	"github.com/mudler/LocalAI/core/backend"
 	"github.com/mudler/LocalAI/core/config"
-	fiberContext "github.com/mudler/LocalAI/core/http/ctx"
+	"github.com/mudler/LocalAI/core/http/middleware"
 	"github.com/mudler/LocalAI/core/schema"
 	"github.com/rs/zerolog/log"
 
 	"github.com/mudler/LocalAI/pkg/model"
 )
 
+// TODO: This is not yet in use. Needs middleware rework, since it is not referenced.
+
 // TokenMetricsEndpoint is an endpoint to get TokensProcessed Per Second for Active SlotID
 //
 //	@Summary	Get TokenMetrics for Active Slot.
@@ -29,18 +31,13 @@ func TokenMetricsEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader,
 			return err
 		}
 
-		modelFile, err := fiberContext.ModelFromContext(c, cl, ml, input.Model, false)
-		if err != nil {
+		modelFile, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_MODEL_NAME).(string)
+		if !ok || modelFile != "" {
 			modelFile = input.Model
 			log.Warn().Msgf("Model not found in context: %s", input.Model)
 		}
 
-		cfg, err := cl.LoadBackendConfigFileByName(modelFile, appConfig.ModelPath,
-			config.LoadOptionDebug(appConfig.Debug),
-			config.LoadOptionThreads(appConfig.Threads),
-			config.LoadOptionContextSize(appConfig.ContextSize),
-			config.LoadOptionF16(appConfig.F16),
-		)
+		cfg, err := cl.LoadBackendConfigFileByNameDefaultOptions(modelFile, appConfig)
 
 		if err != nil {
 			log.Err(err)

core/http/endpoints/localai/tokenize.go

@@ -4,10 +4,9 @@ import (
 	"github.com/gofiber/fiber/v2"
 	"github.com/mudler/LocalAI/core/backend"
 	"github.com/mudler/LocalAI/core/config"
-	fiberContext "github.com/mudler/LocalAI/core/http/ctx"
+	"github.com/mudler/LocalAI/core/http/middleware"
 	"github.com/mudler/LocalAI/core/schema"
 	"github.com/mudler/LocalAI/pkg/model"
-	"github.com/rs/zerolog/log"
 )
 
 // TokenizeEndpoint exposes a REST API to tokenize the content
@@ -16,42 +15,21 @@ import (
 // @Success 200 {object} schema.TokenizeResponse "Response"
 // @Router /v1/tokenize [post]
 func TokenizeEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
-	return func(c *fiber.Ctx) error {
-
-		input := new(schema.TokenizeRequest)
-
-		// Get input data from the request body
-		if err := c.BodyParser(input); err != nil {
-			return err
+	return func(ctx *fiber.Ctx) error {
+		input, ok := ctx.Locals(middleware.CONTEXT_LOCALS_KEY_LOCALAI_REQUEST).(*schema.TokenizeRequest)
+		if !ok || input.Model == "" {
+			return fiber.ErrBadRequest
 		}
 
-		modelFile, err := fiberContext.ModelFromContext(c, cl, ml, input.Model, false)
-		if err != nil {
-			modelFile = input.Model
-			log.Warn().Msgf("Model not found in context: %s", input.Model)
+		cfg, ok := ctx.Locals(middleware.CONTEXT_LOCALS_KEY_MODEL_CONFIG).(*config.BackendConfig)
+		if !ok || cfg == nil {
+			return fiber.ErrBadRequest
 		}
 
-		cfg, err := cl.LoadBackendConfigFileByName(modelFile, appConfig.ModelPath,
-			config.LoadOptionDebug(appConfig.Debug),
-			config.LoadOptionThreads(appConfig.Threads),
-			config.LoadOptionContextSize(appConfig.ContextSize),
-			config.LoadOptionF16(appConfig.F16),
-		)
-
-		if err != nil {
-			log.Err(err)
-			modelFile = input.Model
-			log.Warn().Msgf("Model not found in context: %s", input.Model)
-		} else {
-			modelFile = cfg.Model
-		}
-		log.Debug().Msgf("Request for model: %s", modelFile)
-
 		tokenResponse, err := backend.ModelTokenize(input.Content, ml, *cfg, appConfig)
 		if err != nil {
 			return err
 		}
-
-		return c.JSON(tokenResponse)
+		return ctx.JSON(tokenResponse)
 	}
 }

core/http/endpoints/localai/tts.go

@@ -3,7 +3,7 @@ package localai
 import (
 	"github.com/mudler/LocalAI/core/backend"
 	"github.com/mudler/LocalAI/core/config"
-	fiberContext "github.com/mudler/LocalAI/core/http/ctx"
+	"github.com/mudler/LocalAI/core/http/middleware"
 	"github.com/mudler/LocalAI/pkg/model"
 
 	"github.com/gofiber/fiber/v2"
@@ -24,37 +24,24 @@ import (
 //		@Router		/tts [post]
 func TTSEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
-		input := new(schema.TTSRequest)
-
-		// Get input data from the request body
-		if err := c.BodyParser(input); err != nil {
-			return err
+		input, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_LOCALAI_REQUEST).(*schema.TTSRequest)
+		if !ok || input.Model == "" {
+			return fiber.ErrBadRequest
 		}
 
-		modelFile, err := fiberContext.ModelFromContext(c, cl, ml, input.Model, false)
-		if err != nil {
-			modelFile = input.Model
-			log.Warn().Msgf("Model not found in context: %s", input.Model)
+		cfg, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_MODEL_CONFIG).(*config.BackendConfig)
+		if !ok || cfg == nil {
+			return fiber.ErrBadRequest
 		}
 
-		cfg, err := cl.LoadBackendConfigFileByName(modelFile, appConfig.ModelPath,
-			config.LoadOptionDebug(appConfig.Debug),
-			config.LoadOptionThreads(appConfig.Threads),
-			config.LoadOptionContextSize(appConfig.ContextSize),
-			config.LoadOptionF16(appConfig.F16),
-		)
+		log.Debug().Str("model", input.Model).Msg("LocalAI TTS Request recieved")
 
-		if err != nil {
-			log.Err(err)
-			modelFile = input.Model
-			log.Warn().Msgf("Model not found in context: %s", input.Model)
-		} else {
-			modelFile = cfg.Model
-		}
-		log.Debug().Msgf("Request for model: %s", modelFile)
-
-		if input.Backend != "" {
-			cfg.Backend = input.Backend
+		if cfg.Backend == "" {
+			if input.Backend != "" {
+				cfg.Backend = input.Backend
+			} else {
+				cfg.Backend = model.PiperBackend
+			}
 		}
 
 		if input.Language != "" {
@@ -65,7 +52,7 @@ func TTSEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfi
 			cfg.Voice = input.Voice
 		}
 
-		filePath, _, err := backend.ModelTTS(cfg.Backend, input.Input, modelFile, cfg.Voice, cfg.Language, ml, appConfig, *cfg)
+		filePath, _, err := backend.ModelTTS(input.Input, cfg.Voice, cfg.Language, ml, appConfig, *cfg)
 		if err != nil {
 			return err
 		}

core/http/endpoints/localai/vad.go

@@ -4,9 +4,8 @@ import (
 	"github.com/gofiber/fiber/v2"
 	"github.com/mudler/LocalAI/core/backend"
 	"github.com/mudler/LocalAI/core/config"
-	fiberContext "github.com/mudler/LocalAI/core/http/ctx"
+	"github.com/mudler/LocalAI/core/http/middleware"
 	"github.com/mudler/LocalAI/core/schema"
-	"github.com/mudler/LocalAI/pkg/grpc/proto"
 	"github.com/mudler/LocalAI/pkg/model"
 	"github.com/rs/zerolog/log"
 )
@@ -19,45 +18,20 @@ import (
 // @Router		/vad [post]
 func VADEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
-		input := new(schema.VADRequest)
-
-		// Get input data from the request body
-		if err := c.BodyParser(input); err != nil {
-			return err
+		input, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_LOCALAI_REQUEST).(*schema.VADRequest)
+		if !ok || input.Model == "" {
+			return fiber.ErrBadRequest
 		}
 
-		modelFile, err := fiberContext.ModelFromContext(c, cl, ml, input.Model, false)
-		if err != nil {
-			modelFile = input.Model
-			log.Warn().Msgf("Model not found in context: %s", input.Model)
+		cfg, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_MODEL_CONFIG).(*config.BackendConfig)
+		if !ok || cfg == nil {
+			return fiber.ErrBadRequest
 		}
 
-		cfg, err := cl.LoadBackendConfigFileByName(modelFile, appConfig.ModelPath,
-			config.LoadOptionDebug(appConfig.Debug),
-			config.LoadOptionThreads(appConfig.Threads),
-			config.LoadOptionContextSize(appConfig.ContextSize),
-			config.LoadOptionF16(appConfig.F16),
-		)
+		log.Debug().Str("model", input.Model).Msg("LocalAI VAD Request recieved")
 
-		if err != nil {
-			log.Err(err)
-			modelFile = input.Model
-			log.Warn().Msgf("Model not found in context: %s", input.Model)
-		} else {
-			modelFile = cfg.Model
-		}
-		log.Debug().Msgf("Request for model: %s", modelFile)
+		resp, err := backend.VAD(input, c.Context(), ml, appConfig, *cfg)
 
-		opts := backend.ModelOptions(*cfg, appConfig, model.WithBackendString(cfg.Backend), model.WithModel(modelFile))
-
-		vadModel, err := ml.Load(opts...)
-		if err != nil {
-			return err
-		}
-		req := proto.VADRequest{
-			Audio: input.Audio,
-		}
-		resp, err := vadModel.VAD(c.Context(), &req)
 		if err != nil {
 			return err
 		}

core/http/endpoints/openai/chat.go

@@ -5,18 +5,19 @@ import (
 	"bytes"
 	"encoding/json"
 	"fmt"
-	"strings"
 	"time"
 
 	"github.com/gofiber/fiber/v2"
 	"github.com/google/uuid"
 	"github.com/mudler/LocalAI/core/backend"
 	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/http/middleware"
 	"github.com/mudler/LocalAI/core/schema"
 	"github.com/mudler/LocalAI/pkg/functions"
+
+	"github.com/mudler/LocalAI/pkg/model"
 	"github.com/mudler/LocalAI/pkg/templates"
 
-	model "github.com/mudler/LocalAI/pkg/model"
 	"github.com/rs/zerolog/log"
 	"github.com/valyala/fasthttp"
 )
@@ -174,26 +175,20 @@ func ChatEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, evaluat
 		textContentToReturn = ""
 		id = uuid.New().String()
 		created = int(time.Now().Unix())
-		// Set CorrelationID
-		correlationID := c.Get("X-Correlation-ID")
-		if len(strings.TrimSpace(correlationID)) == 0 {
-			correlationID = id
+		
+		input, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_LOCALAI_REQUEST).(*schema.OpenAIRequest)
+		if !ok || input.Model == "" {
+			return fiber.ErrBadRequest
 		}
-		c.Set("X-Correlation-ID", correlationID)
 
-		// Opt-in extra usage flag
 		extraUsage := c.Get("Extra-Usage", "") != ""
 
-		modelFile, input, err := readRequest(c, cl, ml, startupOptions, true)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
+		config, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_MODEL_CONFIG).(*config.BackendConfig)
+		if !ok || config == nil {
+			return fiber.ErrBadRequest
 		}
 
-		config, input, err := mergeRequestWithConfig(modelFile, input, cl, ml, startupOptions.Debug, startupOptions.Threads, startupOptions.ContextSize, startupOptions.F16)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
-		}
-		log.Debug().Msgf("Configuration read: %+v", config)
+		log.Debug().Msgf("Chat endpoint configuration read: %+v", config)
 
 		funcs := input.Functions
 		shouldUseFn := len(input.Functions) > 0 && config.ShouldUseFunctions()
@@ -543,7 +538,7 @@ func handleQuestion(config *config.BackendConfig, input *schema.OpenAIRequest, m
 		audios = append(audios, m.StringAudios...)
 	}
 
-	predFunc, err := backend.ModelInference(input.Context, prompt, input.Messages, images, videos, audios, ml, *config, o, nil)
+	predFunc, err := backend.ModelInference(input.Context, prompt, input.Messages, images, videos, audios, ml, config, o, nil)
 	if err != nil {
 		log.Error().Err(err).Msg("model inference failed")
 		return "", err

core/http/endpoints/openai/completion.go

@@ -10,12 +10,13 @@ import (
 
 	"github.com/mudler/LocalAI/core/backend"
 	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/http/middleware"
 
 	"github.com/gofiber/fiber/v2"
 	"github.com/google/uuid"
 	"github.com/mudler/LocalAI/core/schema"
 	"github.com/mudler/LocalAI/pkg/functions"
-	model "github.com/mudler/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/pkg/model"
 	"github.com/mudler/LocalAI/pkg/templates"
 	"github.com/rs/zerolog/log"
 	"github.com/valyala/fasthttp"
@@ -27,10 +28,9 @@ import (
 // @Success 200 {object} schema.OpenAIResponse "Response"
 // @Router /v1/completions [post]
 func CompletionEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, evaluator *templates.Evaluator, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
-	id := uuid.New().String()
 	created := int(time.Now().Unix())
 
-	process := func(s string, req *schema.OpenAIRequest, config *config.BackendConfig, loader *model.ModelLoader, responses chan schema.OpenAIResponse, extraUsage bool) {
+	process := func(id string, s string, req *schema.OpenAIRequest, config *config.BackendConfig, loader *model.ModelLoader, responses chan schema.OpenAIResponse, extraUsage bool) {
 		ComputeChoices(req, s, config, appConfig, loader, func(s string, c *[]schema.Choice) {}, func(s string, tokenUsage backend.TokenUsage) bool {
 			usage := schema.OpenAIUsage{
 				PromptTokens:     tokenUsage.Prompt,
@@ -63,22 +63,18 @@ func CompletionEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, e
 	}
 
 	return func(c *fiber.Ctx) error {
-		// Add Correlation
-		c.Set("X-Correlation-ID", id)
-
-		// Opt-in extra usage flag
+		// Handle Correlation
+		id := c.Get("X-Correlation-ID", uuid.New().String())
 		extraUsage := c.Get("Extra-Usage", "") != ""
 
-		modelFile, input, err := readRequest(c, cl, ml, appConfig, true)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
+		input, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_LOCALAI_REQUEST).(*schema.OpenAIRequest)
+		if !ok || input.Model == "" {
+			return fiber.ErrBadRequest
 		}
 
-		log.Debug().Msgf("`input`: %+v", input)
-
-		config, input, err := mergeRequestWithConfig(modelFile, input, cl, ml, appConfig.Debug, appConfig.Threads, appConfig.ContextSize, appConfig.F16)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
+		config, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_MODEL_CONFIG).(*config.BackendConfig)
+		if !ok || config == nil {
+			return fiber.ErrBadRequest
 		}
 
 		if config.ResponseFormatMap != nil {
@@ -122,7 +118,7 @@ func CompletionEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, e
 
 			responses := make(chan schema.OpenAIResponse)
 
-			go process(predInput, input, config, ml, responses, extraUsage)
+			go process(id, predInput, input, config, ml, responses, extraUsage)
 
 			c.Context().SetBodyStreamWriter(fasthttp.StreamWriter(func(w *bufio.Writer) {
 

core/http/endpoints/openai/edit.go

@@ -2,16 +2,17 @@ package openai
 
 import (
 	"encoding/json"
-	"fmt"
 	"time"
 
 	"github.com/mudler/LocalAI/core/backend"
 	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/http/middleware"
 
 	"github.com/gofiber/fiber/v2"
 	"github.com/google/uuid"
 	"github.com/mudler/LocalAI/core/schema"
-	model "github.com/mudler/LocalAI/pkg/model"
+
+	"github.com/mudler/LocalAI/pkg/model"
 	"github.com/mudler/LocalAI/pkg/templates"
 
 	"github.com/rs/zerolog/log"
@@ -25,20 +26,21 @@ import (
 func EditEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, evaluator *templates.Evaluator, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 
 	return func(c *fiber.Ctx) error {
+
+		input, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_LOCALAI_REQUEST).(*schema.OpenAIRequest)
+		if !ok || input.Model == "" {
+			return fiber.ErrBadRequest
+		}
 		// Opt-in extra usage flag
 		extraUsage := c.Get("Extra-Usage", "") != ""
 
-		modelFile, input, err := readRequest(c, cl, ml, appConfig, true)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
+		config, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_MODEL_CONFIG).(*config.BackendConfig)
+		if !ok || config == nil {
+			return fiber.ErrBadRequest
 		}
 
-		config, input, err := mergeRequestWithConfig(modelFile, input, cl, ml, appConfig.Debug, appConfig.Threads, appConfig.ContextSize, appConfig.F16)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
-		}
-
-		log.Debug().Msgf("Parameter Config: %+v", config)
+		log.Debug().Msgf("Edit Endpoint Input : %+v", input)
+		log.Debug().Msgf("Edit Endpoint Config: %+v", *config)
 
 		var result []schema.Choice
 		totalTokenUsage := backend.TokenUsage{}

core/http/endpoints/openai/embeddings.go

@@ -2,11 +2,11 @@ package openai
 
 import (
 	"encoding/json"
-	"fmt"
 	"time"
 
 	"github.com/mudler/LocalAI/core/backend"
 	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/http/middleware"
 	"github.com/mudler/LocalAI/pkg/model"
 
 	"github.com/google/uuid"
@@ -23,14 +23,14 @@ import (
 // @Router /v1/embeddings [post]
 func EmbeddingsEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
-		model, input, err := readRequest(c, cl, ml, appConfig, true)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
+		input, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_LOCALAI_REQUEST).(*schema.OpenAIRequest)
+		if !ok || input.Model == "" {
+			return fiber.ErrBadRequest
 		}
 
-		config, input, err := mergeRequestWithConfig(model, input, cl, ml, appConfig.Debug, appConfig.Threads, appConfig.ContextSize, appConfig.F16)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
+		config, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_MODEL_CONFIG).(*config.BackendConfig)
+		if !ok || config == nil {
+			return fiber.ErrBadRequest
 		}
 
 		log.Debug().Msgf("Parameter Config: %+v", config)

core/http/endpoints/openai/image.go

@@ -15,6 +15,7 @@ import (
 
 	"github.com/google/uuid"
 	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/http/middleware"
 	"github.com/mudler/LocalAI/core/schema"
 
 	"github.com/mudler/LocalAI/core/backend"
@@ -66,25 +67,23 @@ func downloadFile(url string) (string, error) {
 // @Router /v1/images/generations [post]
 func ImageEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
-		m, input, err := readRequest(c, cl, ml, appConfig, false)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
+		input, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_LOCALAI_REQUEST).(*schema.OpenAIRequest)
+		if !ok || input.Model == "" {
+			log.Error().Msg("Image Endpoint - Invalid Input")
+			return fiber.ErrBadRequest
 		}
-
-		if m == "" {
-			m = "stablediffusion"
-		}
-		log.Debug().Msgf("Loading model: %+v", m)
-
-		config, input, err := mergeRequestWithConfig(m, input, cl, ml, appConfig.Debug, 0, 0, false)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
+		
+		config, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_MODEL_CONFIG).(*config.BackendConfig)
+		if !ok || config == nil {
+			log.Error().Msg("Image Endpoint - Invalid Config")
+			return fiber.ErrBadRequest
 		}
 
 		src := ""
 		if input.File != "" {
 
 			fileData := []byte{}
+			var err error
 			// check if input.File is an URL, if so download it and save it
 			// to a temporary file
 			if strings.HasPrefix(input.File, "http://") || strings.HasPrefix(input.File, "https://") {

core/http/endpoints/openai/inference.go

@@ -37,7 +37,7 @@ func ComputeChoices(
 	}
 
 	// get the model function to call for the result
-	predFunc, err := backend.ModelInference(req.Context, predInput, req.Messages, images, videos, audios, loader, *config, o, tokenCallback)
+	predFunc, err := backend.ModelInference(req.Context, predInput, req.Messages, images, videos, audios, loader, config, o, tokenCallback)
 	if err != nil {
 		return result, backend.TokenUsage{}, err
 	}

core/http/endpoints/openai/transcription.go

@@ -1,7 +1,6 @@
 package openai
 
 import (
-	"fmt"
 	"io"
 	"net/http"
 	"os"
@@ -10,6 +9,8 @@ import (
 
 	"github.com/mudler/LocalAI/core/backend"
 	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/http/middleware"
+	"github.com/mudler/LocalAI/core/schema"
 	model "github.com/mudler/LocalAI/pkg/model"
 
 	"github.com/gofiber/fiber/v2"
@@ -25,15 +26,16 @@ import (
 // @Router /v1/audio/transcriptions [post]
 func TranscriptEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
 	return func(c *fiber.Ctx) error {
-		m, input, err := readRequest(c, cl, ml, appConfig, false)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request:%w", err)
+		input, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_LOCALAI_REQUEST).(*schema.OpenAIRequest)
+		if !ok || input.Model == "" {
+			return fiber.ErrBadRequest
 		}
 
-		config, input, err := mergeRequestWithConfig(m, input, cl, ml, appConfig.Debug, appConfig.Threads, appConfig.ContextSize, appConfig.F16)
-		if err != nil {
-			return fmt.Errorf("failed reading parameters from request: %w", err)
+		config, ok := c.Locals(middleware.CONTEXT_LOCALS_KEY_MODEL_CONFIG).(*config.BackendConfig)
+		if !ok || config == nil {
+			return fiber.ErrBadRequest
 		}
+
 		// retrieve the file data from the request
 		file, err := c.FormFile("file")
 		if err != nil {

core/http/middleware/request.go

@@ -1,326 +1,450 @@
-package openai
-
-import (
-	"context"
-	"encoding/json"
-	"fmt"
-	"strconv"
-
-	"github.com/gofiber/fiber/v2"
-	"github.com/google/uuid"
-	"github.com/mudler/LocalAI/core/config"
-	fiberContext "github.com/mudler/LocalAI/core/http/ctx"
-	"github.com/mudler/LocalAI/core/schema"
-	"github.com/mudler/LocalAI/pkg/functions"
-	"github.com/mudler/LocalAI/pkg/model"
-	"github.com/mudler/LocalAI/pkg/templates"
-	"github.com/mudler/LocalAI/pkg/utils"
-	"github.com/rs/zerolog/log"
-)
-
-type correlationIDKeyType string
-
-// CorrelationIDKey to track request across process boundary
-const CorrelationIDKey correlationIDKeyType = "correlationID"
-
-func readRequest(c *fiber.Ctx, cl *config.BackendConfigLoader, ml *model.ModelLoader, o *config.ApplicationConfig, firstModel bool) (string, *schema.OpenAIRequest, error) {
-	input := new(schema.OpenAIRequest)
-
-	// Get input data from the request body
-	if err := c.BodyParser(input); err != nil {
-		return "", nil, fmt.Errorf("failed parsing request body: %w", err)
-	}
-
-	received, _ := json.Marshal(input)
-	// Extract or generate the correlation ID
-	correlationID := c.Get("X-Correlation-ID", uuid.New().String())
-
-	ctx, cancel := context.WithCancel(o.Context)
-	// Add the correlation ID to the new context
-	ctxWithCorrelationID := context.WithValue(ctx, CorrelationIDKey, correlationID)
-
-	input.Context = ctxWithCorrelationID
-	input.Cancel = cancel
-
-	log.Debug().Msgf("Request received: %s", string(received))
-
-	modelFile, err := fiberContext.ModelFromContext(c, cl, ml, input.Model, firstModel)
-
-	return modelFile, input, err
-}
-
-func updateRequestConfig(config *config.BackendConfig, input *schema.OpenAIRequest) {
-	if input.Echo {
-		config.Echo = input.Echo
-	}
-	if input.TopK != nil {
-		config.TopK = input.TopK
-	}
-	if input.TopP != nil {
-		config.TopP = input.TopP
-	}
-
-	if input.Backend != "" {
-		config.Backend = input.Backend
-	}
-
-	if input.ClipSkip != 0 {
-		config.Diffusers.ClipSkip = input.ClipSkip
-	}
-
-	if input.ModelBaseName != "" {
-		config.AutoGPTQ.ModelBaseName = input.ModelBaseName
-	}
-
-	if input.NegativePromptScale != 0 {
-		config.NegativePromptScale = input.NegativePromptScale
-	}
-
-	if input.UseFastTokenizer {
-		config.UseFastTokenizer = input.UseFastTokenizer
-	}
-
-	if input.NegativePrompt != "" {
-		config.NegativePrompt = input.NegativePrompt
-	}
-
-	if input.RopeFreqBase != 0 {
-		config.RopeFreqBase = input.RopeFreqBase
-	}
-
-	if input.RopeFreqScale != 0 {
-		config.RopeFreqScale = input.RopeFreqScale
-	}
-
-	if input.Grammar != "" {
-		config.Grammar = input.Grammar
-	}
-
-	if input.Temperature != nil {
-		config.Temperature = input.Temperature
-	}
-
-	if input.Maxtokens != nil {
-		config.Maxtokens = input.Maxtokens
-	}
-
-	if input.ResponseFormat != nil {
-		switch responseFormat := input.ResponseFormat.(type) {
-		case string:
-			config.ResponseFormat = responseFormat
-		case map[string]interface{}:
-			config.ResponseFormatMap = responseFormat
-		}
-	}
-
-	switch stop := input.Stop.(type) {
-	case string:
-		if stop != "" {
-			config.StopWords = append(config.StopWords, stop)
-		}
-	case []interface{}:
-		for _, pp := range stop {
-			if s, ok := pp.(string); ok {
-				config.StopWords = append(config.StopWords, s)
-			}
-		}
-	}
-
-	if len(input.Tools) > 0 {
-		for _, tool := range input.Tools {
-			input.Functions = append(input.Functions, tool.Function)
-		}
-	}
-
-	if input.ToolsChoice != nil {
-		var toolChoice functions.Tool
-
-		switch content := input.ToolsChoice.(type) {
-		case string:
-			_ = json.Unmarshal([]byte(content), &toolChoice)
-		case map[string]interface{}:
-			dat, _ := json.Marshal(content)
-			_ = json.Unmarshal(dat, &toolChoice)
-		}
-		input.FunctionCall = map[string]interface{}{
-			"name": toolChoice.Function.Name,
-		}
-	}
-
-	// Decode each request's message content
-	imgIndex, vidIndex, audioIndex := 0, 0, 0
-	for i, m := range input.Messages {
-		nrOfImgsInMessage := 0
-		nrOfVideosInMessage := 0
-		nrOfAudiosInMessage := 0
-
-		switch content := m.Content.(type) {
-		case string:
-			input.Messages[i].StringContent = content
-		case []interface{}:
-			dat, _ := json.Marshal(content)
-			c := []schema.Content{}
-			json.Unmarshal(dat, &c)
-
-			textContent := ""
-			// we will template this at the end
-
-		CONTENT:
-			for _, pp := range c {
-				switch pp.Type {
-				case "text":
-					textContent += pp.Text
-					//input.Messages[i].StringContent = pp.Text
-				case "video", "video_url":
-					// Decode content as base64 either if it's an URL or base64 text
-					base64, err := utils.GetContentURIAsBase64(pp.VideoURL.URL)
-					if err != nil {
-						log.Error().Msgf("Failed encoding video: %s", err)
-						continue CONTENT
-					}
-					input.Messages[i].StringVideos = append(input.Messages[i].StringVideos, base64) // TODO: make sure that we only return base64 stuff
-					vidIndex++
-					nrOfVideosInMessage++
-				case "audio_url", "audio":
-					// Decode content as base64 either if it's an URL or base64 text
-					base64, err := utils.GetContentURIAsBase64(pp.AudioURL.URL)
-					if err != nil {
-						log.Error().Msgf("Failed encoding image: %s", err)
-						continue CONTENT
-					}
-					input.Messages[i].StringAudios = append(input.Messages[i].StringAudios, base64) // TODO: make sure that we only return base64 stuff
-					audioIndex++
-					nrOfAudiosInMessage++
-				case "image_url", "image":
-					// Decode content as base64 either if it's an URL or base64 text
-					base64, err := utils.GetContentURIAsBase64(pp.ImageURL.URL)
-					if err != nil {
-						log.Error().Msgf("Failed encoding image: %s", err)
-						continue CONTENT
-					}
-
-					input.Messages[i].StringImages = append(input.Messages[i].StringImages, base64) // TODO: make sure that we only return base64 stuff
-
-					imgIndex++
-					nrOfImgsInMessage++
-				}
-			}
-
-			input.Messages[i].StringContent, _ = templates.TemplateMultiModal(config.TemplateConfig.Multimodal, templates.MultiModalOptions{
-				TotalImages:     imgIndex,
-				TotalVideos:     vidIndex,
-				TotalAudios:     audioIndex,
-				ImagesInMessage: nrOfImgsInMessage,
-				VideosInMessage: nrOfVideosInMessage,
-				AudiosInMessage: nrOfAudiosInMessage,
-			}, textContent)
-		}
-	}
-
-	if input.RepeatPenalty != 0 {
-		config.RepeatPenalty = input.RepeatPenalty
-	}
-
-	if input.FrequencyPenalty != 0 {
-		config.FrequencyPenalty = input.FrequencyPenalty
-	}
-
-	if input.PresencePenalty != 0 {
-		config.PresencePenalty = input.PresencePenalty
-	}
-
-	if input.Keep != 0 {
-		config.Keep = input.Keep
-	}
-
-	if input.Batch != 0 {
-		config.Batch = input.Batch
-	}
-
-	if input.IgnoreEOS {
-		config.IgnoreEOS = input.IgnoreEOS
-	}
-
-	if input.Seed != nil {
-		config.Seed = input.Seed
-	}
-
-	if input.TypicalP != nil {
-		config.TypicalP = input.TypicalP
-	}
-
-	switch inputs := input.Input.(type) {
-	case string:
-		if inputs != "" {
-			config.InputStrings = append(config.InputStrings, inputs)
-		}
-	case []interface{}:
-		for _, pp := range inputs {
-			switch i := pp.(type) {
-			case string:
-				config.InputStrings = append(config.InputStrings, i)
-			case []interface{}:
-				tokens := []int{}
-				for _, ii := range i {
-					tokens = append(tokens, int(ii.(float64)))
-				}
-				config.InputToken = append(config.InputToken, tokens)
-			}
-		}
-	}
-
-	// Can be either a string or an object
-	switch fnc := input.FunctionCall.(type) {
-	case string:
-		if fnc != "" {
-			config.SetFunctionCallString(fnc)
-		}
-	case map[string]interface{}:
-		var name string
-		n, exists := fnc["name"]
-		if exists {
-			nn, e := n.(string)
-			if e {
-				name = nn
-			}
-		}
-		config.SetFunctionCallNameString(name)
-	}
-
-	switch p := input.Prompt.(type) {
-	case string:
-		config.PromptStrings = append(config.PromptStrings, p)
-	case []interface{}:
-		for _, pp := range p {
-			if s, ok := pp.(string); ok {
-				config.PromptStrings = append(config.PromptStrings, s)
-			}
-		}
-	}
-
-	// If a quality was defined as number, convert it to step
-	if input.Quality != "" {
-		q, err := strconv.Atoi(input.Quality)
-		if err == nil {
-			config.Step = q
-		}
-	}
-}
-
-func mergeRequestWithConfig(modelFile string, input *schema.OpenAIRequest, cm *config.BackendConfigLoader, loader *model.ModelLoader, debug bool, threads, ctx int, f16 bool) (*config.BackendConfig, *schema.OpenAIRequest, error) {
-	cfg, err := cm.LoadBackendConfigFileByName(modelFile, loader.ModelPath,
-		config.LoadOptionDebug(debug),
-		config.LoadOptionThreads(threads),
-		config.LoadOptionContextSize(ctx),
-		config.LoadOptionF16(f16),
-	)
-
-	// Set the parameters for the language model prediction
-	updateRequestConfig(cfg, input)
-
-	if !cfg.Validate() {
-		return nil, nil, fmt.Errorf("failed to validate config")
-	}
-
-	return cfg, input, err
-}
+package middleware
+
+import (
+	"context"
+	"encoding/json"
+	"fmt"
+	"strconv"
+	"strings"
+
+	"github.com/google/uuid"
+	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/schema"
+	"github.com/mudler/LocalAI/core/services"
+	"github.com/mudler/LocalAI/pkg/functions"
+	"github.com/mudler/LocalAI/pkg/model"
+	"github.com/mudler/LocalAI/pkg/templates"
+	"github.com/mudler/LocalAI/pkg/utils"
+
+	"github.com/gofiber/fiber/v2"
+	"github.com/rs/zerolog/log"
+)
+
+type correlationIDKeyType string
+
+// CorrelationIDKey to track request across process boundary
+const CorrelationIDKey correlationIDKeyType = "correlationID"
+
+type RequestExtractor struct {
+	backendConfigLoader *config.BackendConfigLoader
+	modelLoader         *model.ModelLoader
+	applicationConfig   *config.ApplicationConfig
+}
+
+func NewRequestExtractor(backendConfigLoader *config.BackendConfigLoader, modelLoader *model.ModelLoader, applicationConfig *config.ApplicationConfig) *RequestExtractor {
+	return &RequestExtractor{
+		backendConfigLoader: backendConfigLoader,
+		modelLoader:         modelLoader,
+		applicationConfig:   applicationConfig,
+	}
+}
+
+const CONTEXT_LOCALS_KEY_MODEL_NAME = "MODEL_NAME"
+const CONTEXT_LOCALS_KEY_LOCALAI_REQUEST = "LOCALAI_REQUEST"
+const CONTEXT_LOCALS_KEY_MODEL_CONFIG = "MODEL_CONFIG"
+
+// TODO: Refactor to not return error if unchanged
+func (re *RequestExtractor) setModelNameFromRequest(ctx *fiber.Ctx) {
+	model, ok := ctx.Locals(CONTEXT_LOCALS_KEY_MODEL_NAME).(string)
+	if ok && model != "" {
+		return
+	}
+	model = ctx.Params("model")
+
+	if (model == "") && ctx.Query("model") != "" {
+		model = ctx.Query("model")
+	}
+
+	if model == "" {
+		// Set model from bearer token, if available
+		bearer := strings.TrimLeft(ctx.Get("authorization"), "Bear ") // "Bearer " => "Bear" to please go-staticcheck. It looks dumb but we might as well take free performance on something called for nearly every request.
+		if bearer != "" {
+			exists, err := services.CheckIfModelExists(re.backendConfigLoader, re.modelLoader, bearer, services.ALWAYS_INCLUDE)
+			if err == nil && exists {
+				model = bearer
+			}
+		}
+	}
+
+	ctx.Locals(CONTEXT_LOCALS_KEY_MODEL_NAME, model)
+}
+
+func (re *RequestExtractor) BuildConstantDefaultModelNameMiddleware(defaultModelName string) fiber.Handler {
+	return func(ctx *fiber.Ctx) error {
+		re.setModelNameFromRequest(ctx)
+		localModelName, ok := ctx.Locals(CONTEXT_LOCALS_KEY_MODEL_NAME).(string)
+		if !ok || localModelName == "" {
+			ctx.Locals(CONTEXT_LOCALS_KEY_MODEL_NAME, defaultModelName)
+			log.Debug().Str("defaultModelName", defaultModelName).Msg("context local model name not found, setting to default")
+		}
+		return ctx.Next()
+	}
+}
+
+func (re *RequestExtractor) BuildFilteredFirstAvailableDefaultModel(filterFn config.BackendConfigFilterFn) fiber.Handler {
+	return func(ctx *fiber.Ctx) error {
+		re.setModelNameFromRequest(ctx)
+		localModelName := ctx.Locals(CONTEXT_LOCALS_KEY_MODEL_NAME).(string)
+		if localModelName != "" { // Don't overwrite existing values
+			return ctx.Next()
+		}
+
+		modelNames, err := services.ListModels(re.backendConfigLoader, re.modelLoader, filterFn, services.SKIP_IF_CONFIGURED)
+		if err != nil {
+			log.Error().Err(err).Msg("non-fatal error calling ListModels during SetDefaultModelNameToFirstAvailable()")
+			return ctx.Next()
+		}
+
+		if len(modelNames) == 0 {
+			log.Warn().Msg("SetDefaultModelNameToFirstAvailable used with no matching models installed")
+			// This is non-fatal - making it so was breaking the case of direct installation of raw models
+			// return errors.New("this endpoint requires at least one model to be installed")
+			return ctx.Next()
+		}
+
+		ctx.Locals(CONTEXT_LOCALS_KEY_MODEL_NAME, modelNames[0])
+		log.Debug().Str("first model name", modelNames[0]).Msg("context local model name not found, setting to the first model")
+		return ctx.Next()
+	}
+}
+
+// TODO: If context and cancel above belong on all methods, move that part of above into here!
+// Otherwise, it's in its own method below for now
+func (re *RequestExtractor) SetModelAndConfig(initializer func() schema.LocalAIRequest) fiber.Handler {
+	return func(ctx *fiber.Ctx) error {
+		input := initializer()
+		if input == nil {
+			return fmt.Errorf("unable to initialize body")
+		}
+		if err := ctx.BodyParser(input); err != nil {
+			return fmt.Errorf("failed parsing request body: %w", err)
+		}
+
+		// If this request doesn't have an associated model name, fetch it from earlier in the middleware chain
+		if input.ModelName(nil) == "" {
+			localModelName, ok := ctx.Locals(CONTEXT_LOCALS_KEY_MODEL_NAME).(string)
+			if ok && localModelName != "" {
+				log.Debug().Str("context localModelName", localModelName).Msg("overriding empty model name in request body with value found earlier in middleware chain")
+				input.ModelName(&localModelName)
+			}
+		}
+
+		cfg, err := re.backendConfigLoader.LoadBackendConfigFileByNameDefaultOptions(input.ModelName(nil), re.applicationConfig)
+
+		if err != nil {
+			log.Err(err)
+			log.Warn().Msgf("Model Configuration File not found for %q", input.ModelName(nil))
+		} else if cfg.Model == "" && input.ModelName(nil) != "" {
+			log.Debug().Str("input.ModelName", input.ModelName(nil)).Msg("config does not include model, using input")
+			cfg.Model = input.ModelName(nil)
+		}
+
+		ctx.Locals(CONTEXT_LOCALS_KEY_LOCALAI_REQUEST, input)
+		ctx.Locals(CONTEXT_LOCALS_KEY_MODEL_CONFIG, cfg)
+
+		return ctx.Next()
+	}
+}
+
+func (re *RequestExtractor) SetOpenAIRequest(ctx *fiber.Ctx) error {
+	input, ok := ctx.Locals(CONTEXT_LOCALS_KEY_LOCALAI_REQUEST).(*schema.OpenAIRequest)
+	if !ok || input.Model == "" {
+		return fiber.ErrBadRequest
+	}
+
+	cfg, ok := ctx.Locals(CONTEXT_LOCALS_KEY_MODEL_CONFIG).(*config.BackendConfig)
+	if !ok || cfg == nil {
+		return fiber.ErrBadRequest
+	}
+
+	// Extract or generate the correlation ID
+	correlationID := ctx.Get("X-Correlation-ID", uuid.New().String())
+	ctx.Set("X-Correlation-ID", correlationID)
+
+	c1, cancel := context.WithCancel(re.applicationConfig.Context)
+	// Add the correlation ID to the new context
+	ctxWithCorrelationID := context.WithValue(c1, CorrelationIDKey, correlationID)
+
+	input.Context = ctxWithCorrelationID
+	input.Cancel = cancel
+
+	err := mergeOpenAIRequestAndBackendConfig(cfg, input)
+	if err != nil {
+		return err
+	}
+
+	if cfg.Model == "" {
+		log.Debug().Str("input.Model", input.Model).Msg("replacing empty cfg.Model with input value")
+		cfg.Model = input.Model
+	}
+
+	ctx.Locals(CONTEXT_LOCALS_KEY_LOCALAI_REQUEST, input)
+	ctx.Locals(CONTEXT_LOCALS_KEY_MODEL_CONFIG, cfg)
+
+	return ctx.Next()
+}
+
+func mergeOpenAIRequestAndBackendConfig(config *config.BackendConfig, input *schema.OpenAIRequest) error {
+	if input.Echo {
+		config.Echo = input.Echo
+	}
+	if input.TopK != nil {
+		config.TopK = input.TopK
+	}
+	if input.TopP != nil {
+		config.TopP = input.TopP
+	}
+
+	if input.Backend != "" {
+		config.Backend = input.Backend
+	}
+
+	if input.ClipSkip != 0 {
+		config.Diffusers.ClipSkip = input.ClipSkip
+	}
+
+	if input.ModelBaseName != "" {
+		config.AutoGPTQ.ModelBaseName = input.ModelBaseName
+	}
+
+	if input.NegativePromptScale != 0 {
+		config.NegativePromptScale = input.NegativePromptScale
+	}
+
+	if input.UseFastTokenizer {
+		config.UseFastTokenizer = input.UseFastTokenizer
+	}
+
+	if input.NegativePrompt != "" {
+		config.NegativePrompt = input.NegativePrompt
+	}
+
+	if input.RopeFreqBase != 0 {
+		config.RopeFreqBase = input.RopeFreqBase
+	}
+
+	if input.RopeFreqScale != 0 {
+		config.RopeFreqScale = input.RopeFreqScale
+	}
+
+	if input.Grammar != "" {
+		config.Grammar = input.Grammar
+	}
+
+	if input.Temperature != nil {
+		config.Temperature = input.Temperature
+	}
+
+	if input.Maxtokens != nil {
+		config.Maxtokens = input.Maxtokens
+	}
+
+	if input.ResponseFormat != nil {
+		switch responseFormat := input.ResponseFormat.(type) {
+		case string:
+			config.ResponseFormat = responseFormat
+		case map[string]interface{}:
+			config.ResponseFormatMap = responseFormat
+		}
+	}
+
+	switch stop := input.Stop.(type) {
+	case string:
+		if stop != "" {
+			config.StopWords = append(config.StopWords, stop)
+		}
+	case []interface{}:
+		for _, pp := range stop {
+			if s, ok := pp.(string); ok {
+				config.StopWords = append(config.StopWords, s)
+			}
+		}
+	}
+
+	if len(input.Tools) > 0 {
+		for _, tool := range input.Tools {
+			input.Functions = append(input.Functions, tool.Function)
+		}
+	}
+
+	if input.ToolsChoice != nil {
+		var toolChoice functions.Tool
+
+		switch content := input.ToolsChoice.(type) {
+		case string:
+			_ = json.Unmarshal([]byte(content), &toolChoice)
+		case map[string]interface{}:
+			dat, _ := json.Marshal(content)
+			_ = json.Unmarshal(dat, &toolChoice)
+		}
+		input.FunctionCall = map[string]interface{}{
+			"name": toolChoice.Function.Name,
+		}
+	}
+
+	// Decode each request's message content
+	imgIndex, vidIndex, audioIndex := 0, 0, 0
+	for i, m := range input.Messages {
+		nrOfImgsInMessage := 0
+		nrOfVideosInMessage := 0
+		nrOfAudiosInMessage := 0
+
+		switch content := m.Content.(type) {
+		case string:
+			input.Messages[i].StringContent = content
+		case []interface{}:
+			dat, _ := json.Marshal(content)
+			c := []schema.Content{}
+			json.Unmarshal(dat, &c)
+
+			textContent := ""
+			// we will template this at the end
+
+		CONTENT:
+			for _, pp := range c {
+				switch pp.Type {
+				case "text":
+					textContent += pp.Text
+					//input.Messages[i].StringContent = pp.Text
+				case "video", "video_url":
+					// Decode content as base64 either if it's an URL or base64 text
+					base64, err := utils.GetContentURIAsBase64(pp.VideoURL.URL)
+					if err != nil {
+						log.Error().Msgf("Failed encoding video: %s", err)
+						continue CONTENT
+					}
+					input.Messages[i].StringVideos = append(input.Messages[i].StringVideos, base64) // TODO: make sure that we only return base64 stuff
+					vidIndex++
+					nrOfVideosInMessage++
+				case "audio_url", "audio":
+					// Decode content as base64 either if it's an URL or base64 text
+					base64, err := utils.GetContentURIAsBase64(pp.AudioURL.URL)
+					if err != nil {
+						log.Error().Msgf("Failed encoding image: %s", err)
+						continue CONTENT
+					}
+					input.Messages[i].StringAudios = append(input.Messages[i].StringAudios, base64) // TODO: make sure that we only return base64 stuff
+					audioIndex++
+					nrOfAudiosInMessage++
+				case "image_url", "image":
+					// Decode content as base64 either if it's an URL or base64 text
+					base64, err := utils.GetContentURIAsBase64(pp.ImageURL.URL)
+					if err != nil {
+						log.Error().Msgf("Failed encoding image: %s", err)
+						continue CONTENT
+					}
+
+					input.Messages[i].StringImages = append(input.Messages[i].StringImages, base64) // TODO: make sure that we only return base64 stuff
+
+					imgIndex++
+					nrOfImgsInMessage++
+				}
+			}
+
+			input.Messages[i].StringContent, _ = templates.TemplateMultiModal(config.TemplateConfig.Multimodal, templates.MultiModalOptions{
+				TotalImages:     imgIndex,
+				TotalVideos:     vidIndex,
+				TotalAudios:     audioIndex,
+				ImagesInMessage: nrOfImgsInMessage,
+				VideosInMessage: nrOfVideosInMessage,
+				AudiosInMessage: nrOfAudiosInMessage,
+			}, textContent)
+		}
+	}
+
+	if input.RepeatPenalty != 0 {
+		config.RepeatPenalty = input.RepeatPenalty
+	}
+
+	if input.FrequencyPenalty != 0 {
+		config.FrequencyPenalty = input.FrequencyPenalty
+	}
+
+	if input.PresencePenalty != 0 {
+		config.PresencePenalty = input.PresencePenalty
+	}
+
+	if input.Keep != 0 {
+		config.Keep = input.Keep
+	}
+
+	if input.Batch != 0 {
+		config.Batch = input.Batch
+	}
+
+	if input.IgnoreEOS {
+		config.IgnoreEOS = input.IgnoreEOS
+	}
+
+	if input.Seed != nil {
+		config.Seed = input.Seed
+	}
+
+	if input.TypicalP != nil {
+		config.TypicalP = input.TypicalP
+	}
+
+	log.Debug().Str("input.Input", fmt.Sprintf("%+v", input.Input))
+
+	switch inputs := input.Input.(type) {
+	case string:
+		if inputs != "" {
+			config.InputStrings = append(config.InputStrings, inputs)
+		}
+	case []interface{}:
+		for _, pp := range inputs {
+			switch i := pp.(type) {
+			case string:
+				config.InputStrings = append(config.InputStrings, i)
+			case []interface{}:
+				tokens := []int{}
+				for _, ii := range i {
+					tokens = append(tokens, int(ii.(float64)))
+				}
+				config.InputToken = append(config.InputToken, tokens)
+			}
+		}
+	}
+
+	// Can be either a string or an object
+	switch fnc := input.FunctionCall.(type) {
+	case string:
+		if fnc != "" {
+			config.SetFunctionCallString(fnc)
+		}
+	case map[string]interface{}:
+		var name string
+		n, exists := fnc["name"]
+		if exists {
+			nn, e := n.(string)
+			if e {
+				name = nn
+			}
+		}
+		config.SetFunctionCallNameString(name)
+	}
+
+	switch p := input.Prompt.(type) {
+	case string:
+		config.PromptStrings = append(config.PromptStrings, p)
+	case []interface{}:
+		for _, pp := range p {
+			if s, ok := pp.(string); ok {
+				config.PromptStrings = append(config.PromptStrings, s)
+			}
+		}
+	}
+
+	// If a quality was defined as number, convert it to step
+	if input.Quality != "" {
+		q, err := strconv.Atoi(input.Quality)
+		if err == nil {
+			config.Step = q
+		}
+	}
+
+	if config.Validate() {
+		return nil
+	}
+	return fmt.Errorf("unable to validate configuration after merging")
+}

core/http/routes/elevenlabs.go

@@ -4,17 +4,26 @@ import (
 	"github.com/gofiber/fiber/v2"
 	"github.com/mudler/LocalAI/core/config"
 	"github.com/mudler/LocalAI/core/http/endpoints/elevenlabs"
+	"github.com/mudler/LocalAI/core/http/middleware"
+	"github.com/mudler/LocalAI/core/schema"
 	"github.com/mudler/LocalAI/pkg/model"
 )
 
 func RegisterElevenLabsRoutes(app *fiber.App,
+	re *middleware.RequestExtractor,
 	cl *config.BackendConfigLoader,
 	ml *model.ModelLoader,
 	appConfig *config.ApplicationConfig) {
 
 	// Elevenlabs
-	app.Post("/v1/text-to-speech/:voice-id", elevenlabs.TTSEndpoint(cl, ml, appConfig))
+	app.Post("/v1/text-to-speech/:voice-id",
+		re.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_TTS)),
+		re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.ElevenLabsTTSRequest) }),
+		elevenlabs.TTSEndpoint(cl, ml, appConfig))
 
-	app.Post("/v1/sound-generation", elevenlabs.SoundGenerationEndpoint(cl, ml, appConfig))
+	app.Post("/v1/sound-generation",
+		re.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_SOUND_GENERATION)),
+		re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.ElevenLabsSoundGenerationRequest) }),
+		elevenlabs.SoundGenerationEndpoint(cl, ml, appConfig))
 
 }

core/http/routes/jina.go

@@ -3,16 +3,22 @@ package routes
 import (
 	"github.com/mudler/LocalAI/core/config"
 	"github.com/mudler/LocalAI/core/http/endpoints/jina"
+	"github.com/mudler/LocalAI/core/http/middleware"
+	"github.com/mudler/LocalAI/core/schema"
 
 	"github.com/gofiber/fiber/v2"
 	"github.com/mudler/LocalAI/pkg/model"
 )
 
 func RegisterJINARoutes(app *fiber.App,
+	re *middleware.RequestExtractor,
 	cl *config.BackendConfigLoader,
 	ml *model.ModelLoader,
 	appConfig *config.ApplicationConfig) {
 
 	// POST endpoint to mimic the reranking
-	app.Post("/v1/rerank", jina.JINARerankEndpoint(cl, ml, appConfig))
+	app.Post("/v1/rerank",
+		re.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_RERANK)),
+		re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.JINARerankRequest) }),
+		jina.JINARerankEndpoint(cl, ml, appConfig))
 }

core/http/routes/localai.go

@@ -5,13 +5,16 @@ import (
 	"github.com/gofiber/swagger"
 	"github.com/mudler/LocalAI/core/config"
 	"github.com/mudler/LocalAI/core/http/endpoints/localai"
+	"github.com/mudler/LocalAI/core/http/middleware"
 	"github.com/mudler/LocalAI/core/p2p"
+	"github.com/mudler/LocalAI/core/schema"
 	"github.com/mudler/LocalAI/core/services"
 	"github.com/mudler/LocalAI/internal"
 	"github.com/mudler/LocalAI/pkg/model"
 )
 
 func RegisterLocalAIRoutes(router *fiber.App,
+	requestExtractor *middleware.RequestExtractor,
 	cl *config.BackendConfigLoader,
 	ml *model.ModelLoader,
 	appConfig *config.ApplicationConfig,
@@ -33,8 +36,18 @@ func RegisterLocalAIRoutes(router *fiber.App,
 		router.Get("/models/jobs", modelGalleryEndpointService.GetAllStatusEndpoint())
 	}
 
-	router.Post("/tts", localai.TTSEndpoint(cl, ml, appConfig))
-	router.Post("/vad", localai.VADEndpoint(cl, ml, appConfig))
+	router.Post("/tts",
+		requestExtractor.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_TTS)),
+		requestExtractor.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.TTSRequest) }),
+		localai.TTSEndpoint(cl, ml, appConfig))
+
+	vadChain := []fiber.Handler{
+		requestExtractor.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_VAD)),
+		requestExtractor.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.VADRequest) }),
+		localai.VADEndpoint(cl, ml, appConfig),
+	}
+	router.Post("/vad", vadChain...)
+	router.Post("/v1/vad", vadChain...)
 
 	// Stores
 	sl := model.NewModelLoader("")
@@ -47,10 +60,14 @@ func RegisterLocalAIRoutes(router *fiber.App,
 		router.Get("/metrics", localai.LocalAIMetricsEndpoint())
 	}
 
-	// Experimental Backend Statistics Module
+	// Backend Statistics Module
+	// TODO: Should these use standard middlewares? Refactor later, they are extremely simple.
 	backendMonitorService := services.NewBackendMonitorService(ml, cl, appConfig) // Split out for now
 	router.Get("/backend/monitor", localai.BackendMonitorEndpoint(backendMonitorService))
 	router.Post("/backend/shutdown", localai.BackendShutdownEndpoint(backendMonitorService))
+	// The v1/* urls are exactly the same as above - makes local e2e testing easier if they are registered.
+	router.Get("/v1/backend/monitor", localai.BackendMonitorEndpoint(backendMonitorService))
+	router.Post("/v1/backend/shutdown", localai.BackendShutdownEndpoint(backendMonitorService))
 
 	// p2p
 	if p2p.IsP2PEnabled() {
@@ -67,6 +84,9 @@ func RegisterLocalAIRoutes(router *fiber.App,
 	router.Get("/system", localai.SystemInformations(ml, appConfig))
 
 	// misc
-	router.Post("/v1/tokenize", localai.TokenizeEndpoint(cl, ml, appConfig))
+	router.Post("/v1/tokenize",
+		requestExtractor.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_TOKENIZE)),
+		requestExtractor.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.TokenizeRequest) }),
+		localai.TokenizeEndpoint(cl, ml, appConfig))
 
 }

core/http/routes/openai.go

@@ -3,51 +3,50 @@ package routes
 import (
 	"github.com/gofiber/fiber/v2"
 	"github.com/mudler/LocalAI/core/application"
+	"github.com/mudler/LocalAI/core/config"
 	"github.com/mudler/LocalAI/core/http/endpoints/localai"
 	"github.com/mudler/LocalAI/core/http/endpoints/openai"
+	"github.com/mudler/LocalAI/core/http/middleware"
+	"github.com/mudler/LocalAI/core/schema"
 )
 
 func RegisterOpenAIRoutes(app *fiber.App,
+	re *middleware.RequestExtractor,
 	application *application.Application) {
 	// openAI compatible API endpoint
 
 	// chat
-	app.Post("/v1/chat/completions",
-		openai.ChatEndpoint(
-			application.BackendLoader(),
-			application.ModelLoader(),
-			application.TemplatesEvaluator(),
-			application.ApplicationConfig(),
-		),
-	)
-
-	app.Post("/chat/completions",
-		openai.ChatEndpoint(
-			application.BackendLoader(),
-			application.ModelLoader(),
-			application.TemplatesEvaluator(),
-			application.ApplicationConfig(),
-		),
-	)
+	chatChain := []fiber.Handler{
+		re.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_CHAT)),
+		re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.OpenAIRequest) }),
+		re.SetOpenAIRequest,
+		openai.ChatEndpoint(application.BackendLoader(), application.ModelLoader(), application.TemplatesEvaluator(), application.ApplicationConfig()),
+	}
+	app.Post("/v1/chat/completions", chatChain...)
+	app.Post("/chat/completions", chatChain...)
 
 	// edit
-	app.Post("/v1/edits",
-		openai.EditEndpoint(
-			application.BackendLoader(),
-			application.ModelLoader(),
-			application.TemplatesEvaluator(),
-			application.ApplicationConfig(),
-		),
-	)
+	editChain := []fiber.Handler{
+		re.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_EDIT)),
+		re.BuildConstantDefaultModelNameMiddleware("gpt-4o"),
+		re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.OpenAIRequest) }),
+		re.SetOpenAIRequest,
+		openai.EditEndpoint(application.BackendLoader(), application.ModelLoader(), application.TemplatesEvaluator(), application.ApplicationConfig()),
+	}
+	app.Post("/v1/edits", editChain...)
+	app.Post("/edits", editChain...)
 
-	app.Post("/edits",
-		openai.EditEndpoint(
-			application.BackendLoader(),
-			application.ModelLoader(),
-			application.TemplatesEvaluator(),
-			application.ApplicationConfig(),
-		),
-	)
+	// completion
+	completionChain := []fiber.Handler{
+		re.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_COMPLETION)),
+		re.BuildConstantDefaultModelNameMiddleware("gpt-4o"),
+		re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.OpenAIRequest) }),
+		re.SetOpenAIRequest,
+		openai.CompletionEndpoint(application.BackendLoader(), application.ModelLoader(), application.TemplatesEvaluator(), application.ApplicationConfig()),
+	}
+	app.Post("/v1/completions", completionChain...)
+	app.Post("/completions", completionChain...)
+	app.Post("/v1/engines/:model/completions", completionChain...)
 
 	// assistant
 	app.Get("/v1/assistants", openai.ListAssistantsEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))
@@ -81,53 +80,37 @@ func RegisterOpenAIRoutes(app *fiber.App,
 	app.Get("/v1/files/:file_id/content", openai.GetFilesContentsEndpoint(application.BackendLoader(), application.ApplicationConfig()))
 	app.Get("/files/:file_id/content", openai.GetFilesContentsEndpoint(application.BackendLoader(), application.ApplicationConfig()))
 
-	// completion
-	app.Post("/v1/completions",
-		openai.CompletionEndpoint(
-			application.BackendLoader(),
-			application.ModelLoader(),
-			application.TemplatesEvaluator(),
-			application.ApplicationConfig(),
-		),
-	)
-
-	app.Post("/completions",
-		openai.CompletionEndpoint(
-			application.BackendLoader(),
-			application.ModelLoader(),
-			application.TemplatesEvaluator(),
-			application.ApplicationConfig(),
-		),
-	)
-
-	app.Post("/v1/engines/:model/completions",
-		openai.CompletionEndpoint(
-			application.BackendLoader(),
-			application.ModelLoader(),
-			application.TemplatesEvaluator(),
-			application.ApplicationConfig(),
-		),
-	)
-
 	// embeddings
-	app.Post("/v1/embeddings", openai.EmbeddingsEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))
-	app.Post("/embeddings", openai.EmbeddingsEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))
-	app.Post("/v1/engines/:model/embeddings", openai.EmbeddingsEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))
+	embeddingChain := []fiber.Handler{
+		re.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_EMBEDDINGS)),
+		re.BuildConstantDefaultModelNameMiddleware("gpt-4o"),
+		re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.OpenAIRequest) }),
+		re.SetOpenAIRequest,
+		openai.EmbeddingsEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()),
+	}
+	app.Post("/v1/embeddings", embeddingChain...)
+	app.Post("/embeddings", embeddingChain...)
+	app.Post("/v1/engines/:model/embeddings", embeddingChain...)
 
 	// audio
-	app.Post("/v1/audio/transcriptions", openai.TranscriptEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))
-	app.Post("/v1/audio/speech", localai.TTSEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))
+	app.Post("/v1/audio/transcriptions",
+		re.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_TRANSCRIPT)),
+		re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.OpenAIRequest) }),
+		re.SetOpenAIRequest,
+		openai.TranscriptEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()),
+	)
+
+	app.Post("/v1/audio/speech",
+		re.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_TTS)),
+		re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.TTSRequest) }),
+		localai.TTSEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))
 
 	// images
-	app.Post("/v1/images/generations", openai.ImageEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))
-
-	if application.ApplicationConfig().ImageDir != "" {
-		app.Static("/generated-images", application.ApplicationConfig().ImageDir)
-	}
-
-	if application.ApplicationConfig().AudioDir != "" {
-		app.Static("/generated-audio", application.ApplicationConfig().AudioDir)
-	}
+	app.Post("/v1/images/generations",
+		re.BuildConstantDefaultModelNameMiddleware("stablediffusion"),
+		re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.OpenAIRequest) }),
+		re.SetOpenAIRequest,
+		openai.ImageEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))
 
 	// List models
 	app.Get("/v1/models", openai.ListModelsEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))

core/http/routes/ui.go

@@ -3,7 +3,9 @@ package routes
 import (
 	"fmt"
 	"html/template"
+	"math"
 	"sort"
+	"strconv"
 	"strings"
 
 	"github.com/mudler/LocalAI/core/config"
@@ -126,6 +128,8 @@ func RegisterUIRoutes(app *fiber.App,
 		// Show the Models page (all models)
 		app.Get("/browse", func(c *fiber.Ctx) error {
 			term := c.Query("term")
+			page := c.Query("page")
+			items := c.Query("items")
 
 			models, _ := gallery.AvailableGalleryModels(appConfig.Galleries, appConfig.ModelPath)
 
@@ -164,6 +168,47 @@ func RegisterUIRoutes(app *fiber.App,
 				//	"ApplicationConfig": appConfig,
 			}
 
+			if page == "" {
+				page = "1"
+			}
+
+			if page != "" {
+				// return a subset of the models
+				pageNum, err := strconv.Atoi(page)
+				if err != nil {
+					return c.Status(fiber.StatusBadRequest).SendString("Invalid page number")
+				}
+
+				if pageNum == 0 {
+					return c.Render("views/models", summary)
+				}
+
+				itemsNum, err := strconv.Atoi(items)
+				if err != nil {
+					itemsNum = 21
+				}
+
+				totalPages := int(math.Ceil(float64(len(models)) / float64(itemsNum)))
+
+				models = models.Paginate(pageNum, itemsNum)
+
+				prevPage := pageNum - 1
+				nextPage := pageNum + 1
+				if prevPage < 1 {
+					prevPage = 1
+				}
+				if nextPage > totalPages {
+					nextPage = totalPages
+				}
+				if prevPage != pageNum {
+					summary["PrevPage"] = prevPage
+				}
+				summary["NextPage"] = nextPage
+				summary["TotalPages"] = totalPages
+				summary["CurrentPage"] = pageNum
+				summary["Models"] = template.HTML(elements.ListModels(models, processingModels, galleryService))
+			}
+
 			// Render index
 			return c.Render("views/models", summary)
 		})
@@ -171,6 +216,9 @@ func RegisterUIRoutes(app *fiber.App,
 		// Show the models, filtered from the user input
 		// https://htmx.org/examples/active-search/
 		app.Post("/browse/search/models", func(c *fiber.Ctx) error {
+			page := c.Query("page")
+			items := c.Query("items")
+
 			form := struct {
 				Search string `form:"search"`
 			}{}
@@ -180,7 +228,26 @@ func RegisterUIRoutes(app *fiber.App,
 
 			models, _ := gallery.AvailableGalleryModels(appConfig.Galleries, appConfig.ModelPath)
 
-			return c.SendString(elements.ListModels(gallery.GalleryModels(models).Search(form.Search), processingModels, galleryService))
+			if page != "" {
+				// return a subset of the models
+				pageNum, err := strconv.Atoi(page)
+				if err != nil {
+					return c.Status(fiber.StatusBadRequest).SendString("Invalid page number")
+				}
+
+				itemsNum, err := strconv.Atoi(items)
+				if err != nil {
+					itemsNum = 21
+				}
+
+				models = models.Paginate(pageNum, itemsNum)
+			}
+
+			if form.Search != "" {
+				models = models.Search(form.Search)
+			}
+
+			return c.SendString(elements.ListModels(models, processingModels, galleryService))
 		})
 
 		/*
@@ -305,23 +372,6 @@ func RegisterUIRoutes(app *fiber.App,
 		})
 	}
 
-	// Show the Chat page
-	app.Get("/chat/:model", func(c *fiber.Ctx) error {
-		backendConfigs, _ := services.ListModels(cl, ml, config.NoFilterFn, services.SKIP_IF_CONFIGURED)
-
-		summary := fiber.Map{
-			"Title":        "LocalAI - Chat with " + c.Params("model"),
-			"BaseURL":      utils.BaseURL(c),
-			"ModelsConfig": backendConfigs,
-			"Model":        c.Params("model"),
-			"Version":      internal.PrintableVersion(),
-			"IsP2PEnabled": p2p.IsP2PEnabled(),
-		}
-
-		// Render index
-		return c.Render("views/chat", summary)
-	})
-
 	app.Get("/talk/", func(c *fiber.Ctx) error {
 		backendConfigs, _ := services.ListModels(cl, ml, config.NoFilterFn, services.SKIP_IF_CONFIGURED)
 
@@ -344,21 +394,73 @@ func RegisterUIRoutes(app *fiber.App,
 	})
 
 	app.Get("/chat/", func(c *fiber.Ctx) error {
+		backendConfigs := cl.GetAllBackendConfigs()
+		modelsWithoutConfig, _ := services.ListModels(cl, ml, config.NoFilterFn, services.LOOSE_ONLY)
 
-		backendConfigs, _ := services.ListModels(cl, ml, config.NoFilterFn, services.SKIP_IF_CONFIGURED)
-
-		if len(backendConfigs) == 0 {
+		if len(backendConfigs)+len(modelsWithoutConfig) == 0 {
 			// If no model is available redirect to the index which suggests how to install models
 			return c.Redirect(utils.BaseURL(c))
 		}
+		modelThatCanBeUsed := ""
+		galleryConfigs := map[string]*gallery.Config{}
+
+		for _, m := range backendConfigs {
+			cfg, err := gallery.GetLocalModelConfiguration(ml.ModelPath, m.Name)
+			if err != nil {
+				continue
+			}
+			galleryConfigs[m.Name] = cfg
+		}
+
+		title := "LocalAI - Chat"
+
+		for _, b := range backendConfigs {
+			if b.HasUsecases(config.FLAG_CHAT) {
+				modelThatCanBeUsed = b.Name
+				title = "LocalAI - Chat with " + modelThatCanBeUsed
+				break
+			}
+		}
 
 		summary := fiber.Map{
-			"Title":        "LocalAI - Chat with " + backendConfigs[0],
-			"BaseURL":      utils.BaseURL(c),
-			"ModelsConfig": backendConfigs,
-			"Model":        backendConfigs[0],
-			"Version":      internal.PrintableVersion(),
-			"IsP2PEnabled": p2p.IsP2PEnabled(),
+			"Title":               title,
+			"BaseURL":             utils.BaseURL(c),
+			"ModelsWithoutConfig": modelsWithoutConfig,
+			"GalleryConfig":       galleryConfigs,
+			"ModelsConfig":        backendConfigs,
+			"Model":               modelThatCanBeUsed,
+			"Version":             internal.PrintableVersion(),
+			"IsP2PEnabled":        p2p.IsP2PEnabled(),
+		}
+
+		// Render index
+		return c.Render("views/chat", summary)
+	})
+
+	// Show the Chat page
+	app.Get("/chat/:model", func(c *fiber.Ctx) error {
+		backendConfigs := cl.GetAllBackendConfigs()
+		modelsWithoutConfig, _ := services.ListModels(cl, ml, config.NoFilterFn, services.LOOSE_ONLY)
+
+		galleryConfigs := map[string]*gallery.Config{}
+
+		for _, m := range backendConfigs {
+			cfg, err := gallery.GetLocalModelConfiguration(ml.ModelPath, m.Name)
+			if err != nil {
+				continue
+			}
+			galleryConfigs[m.Name] = cfg
+		}
+
+		summary := fiber.Map{
+			"Title":               "LocalAI - Chat with " + c.Params("model"),
+			"BaseURL":             utils.BaseURL(c),
+			"ModelsConfig":        backendConfigs,
+			"GalleryConfig":       galleryConfigs,
+			"ModelsWithoutConfig": modelsWithoutConfig,
+			"Model":               c.Params("model"),
+			"Version":             internal.PrintableVersion(),
+			"IsP2PEnabled":        p2p.IsP2PEnabled(),
 		}
 
 		// Render index
@@ -367,14 +469,16 @@ func RegisterUIRoutes(app *fiber.App,
 
 	app.Get("/text2image/:model", func(c *fiber.Ctx) error {
 		backendConfigs := cl.GetAllBackendConfigs()
+		modelsWithoutConfig, _ := services.ListModels(cl, ml, config.NoFilterFn, services.LOOSE_ONLY)
 
 		summary := fiber.Map{
-			"Title":        "LocalAI - Generate images with " + c.Params("model"),
-			"BaseURL":      utils.BaseURL(c),
-			"ModelsConfig": backendConfigs,
-			"Model":        c.Params("model"),
-			"Version":      internal.PrintableVersion(),
-			"IsP2PEnabled": p2p.IsP2PEnabled(),
+			"Title":               "LocalAI - Generate images with " + c.Params("model"),
+			"BaseURL":             utils.BaseURL(c),
+			"ModelsConfig":        backendConfigs,
+			"ModelsWithoutConfig": modelsWithoutConfig,
+			"Model":               c.Params("model"),
+			"Version":             internal.PrintableVersion(),
+			"IsP2PEnabled":        p2p.IsP2PEnabled(),
 		}
 
 		// Render index
@@ -382,21 +486,33 @@ func RegisterUIRoutes(app *fiber.App,
 	})
 
 	app.Get("/text2image/", func(c *fiber.Ctx) error {
-
 		backendConfigs := cl.GetAllBackendConfigs()
+		modelsWithoutConfig, _ := services.ListModels(cl, ml, config.NoFilterFn, services.LOOSE_ONLY)
 
-		if len(backendConfigs) == 0 {
+		if len(backendConfigs)+len(modelsWithoutConfig) == 0 {
 			// If no model is available redirect to the index which suggests how to install models
 			return c.Redirect(utils.BaseURL(c))
 		}
 
+		modelThatCanBeUsed := ""
+		title := "LocalAI - Generate images"
+
+		for _, b := range backendConfigs {
+			if b.HasUsecases(config.FLAG_IMAGE) {
+				modelThatCanBeUsed = b.Name
+				title = "LocalAI - Generate images with " + modelThatCanBeUsed
+				break
+			}
+		}
+
 		summary := fiber.Map{
-			"Title":        "LocalAI - Generate images with " + backendConfigs[0].Name,
-			"BaseURL":      utils.BaseURL(c),
-			"ModelsConfig": backendConfigs,
-			"Model":        backendConfigs[0].Name,
-			"Version":      internal.PrintableVersion(),
-			"IsP2PEnabled": p2p.IsP2PEnabled(),
+			"Title":               title,
+			"BaseURL":             utils.BaseURL(c),
+			"ModelsConfig":        backendConfigs,
+			"ModelsWithoutConfig": modelsWithoutConfig,
+			"Model":               modelThatCanBeUsed,
+			"Version":             internal.PrintableVersion(),
+			"IsP2PEnabled":        p2p.IsP2PEnabled(),
 		}
 
 		// Render index
@@ -405,14 +521,16 @@ func RegisterUIRoutes(app *fiber.App,
 
 	app.Get("/tts/:model", func(c *fiber.Ctx) error {
 		backendConfigs := cl.GetAllBackendConfigs()
+		modelsWithoutConfig, _ := services.ListModels(cl, ml, config.NoFilterFn, services.LOOSE_ONLY)
 
 		summary := fiber.Map{
-			"Title":        "LocalAI - Generate images with " + c.Params("model"),
-			"BaseURL":      utils.BaseURL(c),
-			"ModelsConfig": backendConfigs,
-			"Model":        c.Params("model"),
-			"Version":      internal.PrintableVersion(),
-			"IsP2PEnabled": p2p.IsP2PEnabled(),
+			"Title":               "LocalAI - Generate images with " + c.Params("model"),
+			"BaseURL":             utils.BaseURL(c),
+			"ModelsConfig":        backendConfigs,
+			"ModelsWithoutConfig": modelsWithoutConfig,
+			"Model":               c.Params("model"),
+			"Version":             internal.PrintableVersion(),
+			"IsP2PEnabled":        p2p.IsP2PEnabled(),
 		}
 
 		// Render index
@@ -420,21 +538,32 @@ func RegisterUIRoutes(app *fiber.App,
 	})
 
 	app.Get("/tts/", func(c *fiber.Ctx) error {
-
 		backendConfigs := cl.GetAllBackendConfigs()
+		modelsWithoutConfig, _ := services.ListModels(cl, ml, config.NoFilterFn, services.LOOSE_ONLY)
 
-		if len(backendConfigs) == 0 {
+		if len(backendConfigs)+len(modelsWithoutConfig) == 0 {
 			// If no model is available redirect to the index which suggests how to install models
 			return c.Redirect(utils.BaseURL(c))
 		}
 
+		modelThatCanBeUsed := ""
+		title := "LocalAI - Generate audio"
+
+		for _, b := range backendConfigs {
+			if b.HasUsecases(config.FLAG_TTS) {
+				modelThatCanBeUsed = b.Name
+				title = "LocalAI - Generate audio with " + modelThatCanBeUsed
+				break
+			}
+		}
 		summary := fiber.Map{
-			"Title":        "LocalAI - Generate audio with " + backendConfigs[0].Name,
-			"BaseURL":      utils.BaseURL(c),
-			"ModelsConfig": backendConfigs,
-			"Model":        backendConfigs[0].Name,
-			"IsP2PEnabled": p2p.IsP2PEnabled(),
-			"Version":      internal.PrintableVersion(),
+			"Title":               title,
+			"BaseURL":             utils.BaseURL(c),
+			"ModelsConfig":        backendConfigs,
+			"ModelsWithoutConfig": modelsWithoutConfig,
+			"Model":               modelThatCanBeUsed,
+			"IsP2PEnabled":        p2p.IsP2PEnabled(),
+			"Version":             internal.PrintableVersion(),
 		}
 
 		// Render index

core/http/static/chat.js

@@ -27,10 +27,19 @@ SOFTWARE.
 
 */
 
-function submitKey(event) {
-    event.preventDefault();
-    localStorage.setItem("key", document.getElementById("apiKey").value);
-    document.getElementById("apiKey").blur();
+function toggleLoader(show) {
+  const loader = document.getElementById('loader');
+  const sendButton = document.getElementById('send-button');
+  
+  if (show) {
+    loader.style.display = 'block';
+    sendButton.style.display = 'none';
+    document.getElementById("input").disabled = true;
+  } else {
+    document.getElementById("input").disabled = false;
+    loader.style.display = 'none';
+    sendButton.style.display = 'block';
+  }
 }
 
 function submitSystemPrompt(event) {
@@ -47,10 +56,9 @@ function submitPrompt(event) {
   const input = document.getElementById("input").value;
   Alpine.store("chat").add("user", input, image);
   document.getElementById("input").value = "";
-  const key = localStorage.getItem("key");
   const systemPrompt = localStorage.getItem("system_prompt");
-
-  promptGPT(systemPrompt, key, input);
+  Alpine.nextTick(() => { document.getElementById('messages').scrollIntoView(false); });
+  promptGPT(systemPrompt, input);
 }
 
 function readInputImage() {
@@ -67,14 +75,13 @@ function readInputImage() {
 }
 
 
-  async function promptGPT(systemPrompt, key, input) {
+  async function promptGPT(systemPrompt, input) {
     const model = document.getElementById("chat-model").value;
     // Set class "loader" to the element with "loader" id
     //document.getElementById("loader").classList.add("loader");
     // Make the "loader" visible
-    document.getElementById("loader").style.display = "block";
-    document.getElementById("input").disabled = true;
-    document.getElementById('messages').scrollIntoView(false)
+    toggleLoader(true);
+
 
     messages = Alpine.store("chat").messages();
 
@@ -146,7 +153,6 @@ function readInputImage() {
     const response = await fetch("v1/chat/completions", {
       method: "POST",
       headers: {
-        Authorization: `Bearer ${key}`,
         "Content-Type": "application/json",
       },
       body: JSON.stringify({
@@ -181,8 +187,8 @@ function readInputImage() {
       const chatStore = Alpine.store("chat");
       chatStore.add("assistant", token);
       // Efficiently scroll into view without triggering multiple reflows
-      const messages = document.getElementById('messages');
-      messages.scrollTop = messages.scrollHeight;
+      // const messages = document.getElementById('messages');
+      // messages.scrollTop = messages.scrollHeight;
     };
 
     let buffer = "";
@@ -244,30 +250,20 @@ function readInputImage() {
     }
 
     // Remove class "loader" from the element with "loader" id
-    //document.getElementById("loader").classList.remove("loader");
-    document.getElementById("loader").style.display = "none";
-    // enable input
-    document.getElementById("input").disabled = false;
+    toggleLoader(false);
+
     // scroll to the bottom of the chat
     document.getElementById('messages').scrollIntoView(false)
     // set focus to the input
     document.getElementById("input").focus();
   }
 
-  document.getElementById("key").addEventListener("submit", submitKey);
   document.getElementById("system_prompt").addEventListener("submit", submitSystemPrompt);
 
   document.getElementById("prompt").addEventListener("submit", submitPrompt);
   document.getElementById("input").focus();
   document.getElementById("input_image").addEventListener("change", readInputImage);
 
-  storeKey = localStorage.getItem("key");
-  if (storeKey) {
-    document.getElementById("apiKey").value = storeKey;
-  } else {
-    document.getElementById("apiKey").value = null;
-  }
-
   storesystemPrompt = localStorage.getItem("system_prompt");
   if (storesystemPrompt) {
     document.getElementById("systemPrompt").value = storesystemPrompt;

core/http/static/general.css

@@ -10,18 +10,6 @@ body {
 .htmx-request .htmx-indicator{
     opacity:1
 }
-/* Loader (https://cssloaders.github.io/) */
-.loader {
-  width: 12px;
-  height: 12px;
-  border-radius: 50%;
-  display: block;
-  margin:15px auto;
-  position: relative;
-  color: #FFF;
-  box-sizing: border-box;
-  animation: animloader 2s linear infinite;
-}
 
 @keyframes animloader {
   0% { box-shadow: 14px 0 0 -2px,  38px 0 0 -2px,  -14px 0 0 -2px,  -38px 0 0 -2px; }

core/http/static/image.js

@@ -1,48 +1,11 @@
-/*
-
-https://github.com/david-haerer/chatapi
-
-MIT License
-
-Copyright (c) 2023 David Härer
-Copyright (c) 2024 Ettore Di Giacinto
-
-Permission is hereby granted, free of charge, to any person obtaining a copy
-of this software and associated documentation files (the "Software"), to deal
-in the Software without restriction, including without limitation the rights
-to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
-copies of the Software, and to permit persons to whom the Software is
-furnished to do so, subject to the following conditions:
-
-The above copyright notice and this permission notice shall be included in all
-copies or substantial portions of the Software.
-
-THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-SOFTWARE.
-
-*/
-function submitKey(event) {
-    event.preventDefault();
-    localStorage.setItem("key", document.getElementById("apiKey").value);
-    document.getElementById("apiKey").blur();
-  }
-  
-
 function genImage(event) {
   event.preventDefault();
   const input = document.getElementById("input").value;
-  const key = localStorage.getItem("key");
-
-  promptDallE(key, input);
 
+  promptDallE(input);
 }
   
-async function promptDallE(key, input) {
+async function promptDallE(input) {
   document.getElementById("loader").style.display = "block";
   document.getElementById("input").value = "";
   document.getElementById("input").disabled = true;
@@ -51,7 +14,6 @@ async function promptDallE(key, input) {
   const response = await fetch("v1/images/generations", {
     method: "POST",
     headers: {
-      Authorization: `Bearer ${key}`,
       "Content-Type": "application/json",
     },
     body: JSON.stringify({
@@ -84,13 +46,6 @@ async function promptDallE(key, input) {
   document.getElementById("input").focus();
 }
 
-document.getElementById("key").addEventListener("submit", submitKey);
 document.getElementById("input").focus();
 document.getElementById("genimage").addEventListener("submit", genImage);
 document.getElementById("loader").style.display = "none";
-
-const storeKey = localStorage.getItem("key");
-if (storeKey) {
-  document.getElementById("apiKey").value = storeKey;
-}
-

core/http/static/talk.js

@@ -9,10 +9,6 @@ let isRecording = false;
 let conversationHistory = [];
 let resetTimer;
 
-function getApiKey() {
-    return document.getElementById('apiKey').value;
-}
-
 function getModel() {
     return document.getElementById('modelSelect').value;
 }
@@ -99,34 +95,13 @@ function stopRecording() {
     };
 }
 
-function submitKey(event) {
-    event.preventDefault();
-    localStorage.setItem("key", document.getElementById("apiKey").value);
-    document.getElementById("apiKey").blur();
-}
-
-document.getElementById("key").addEventListener("submit", submitKey);
-
-
-storeKey = localStorage.getItem("key");
-if (storeKey) {
-  document.getElementById("apiKey").value = storeKey;
-} else {
-  document.getElementById("apiKey").value = null;
-}
-
-
 async function sendAudioToWhisper(audioBlob) {
     const formData = new FormData();
     formData.append('file', audioBlob);
     formData.append('model', getWhisperModel());
-    API_KEY = localStorage.getItem("key");
 
     const response = await fetch('v1/audio/transcriptions', {
         method: 'POST',
-        headers: {
-            'Authorization': `Bearer ${API_KEY}`
-        },
         body: formData
     });
 
@@ -137,14 +112,9 @@ async function sendAudioToWhisper(audioBlob) {
 
 async function sendTextToChatGPT(text) {
     conversationHistory.push({ role: "user", content: text });
-    API_KEY = localStorage.getItem("key");
 
     const response = await fetch('v1/chat/completions', {
         method: 'POST',
-        headers: {
-            'Authorization': `Bearer ${API_KEY}`,
-            'Content-Type': 'application/json'
-        },
         body: JSON.stringify({
             model: getModel(),
             messages: conversationHistory
@@ -161,13 +131,10 @@ async function sendTextToChatGPT(text) {
 }
 
 async function getTextToSpeechAudio(text) {
-    API_KEY = localStorage.getItem("key");
-
     const response = await fetch('v1/audio/speech', {
         
         method: 'POST',
         headers: {
-            'Authorization': `Bearer ${API_KEY}`,
             'Content-Type': 'application/json'
         },
         body: JSON.stringify({ 

core/http/static/tts.js

@@ -1,64 +1,204 @@
-function submitKey(event) {
-    event.preventDefault();
-    localStorage.setItem("key", document.getElementById("apiKey").value);
-    document.getElementById("apiKey").blur();
-  }
-  
+// Initialize Alpine store for API key management
+document.addEventListener('alpine:init', () => {
+  Alpine.store('chat', {  });
+});
 
 function genAudio(event) {
   event.preventDefault();
   const input = document.getElementById("input").value;
-  const key = localStorage.getItem("key");
 
-  tts(key, input);
-}
-  
-async function tts(key, input) {
-  document.getElementById("loader").style.display = "block";
-  document.getElementById("input").value = "";
-  document.getElementById("input").disabled = true;
-
-  const model = document.getElementById("tts-model").value;
-  const response = await fetch("tts", {
-    method: "POST",
-    headers: {
-      Authorization: `Bearer ${key}`,
-      "Content-Type": "application/json",
-    },
-    body: JSON.stringify({
-      model: model,
-      input: input,
-    }),
-  });
-  if (!response.ok) {
-    const jsonData = await response.json(); // Now safely parse JSON
-    var div = document.getElementById('result');
-    div.innerHTML = '<p style="color:red;">Error: ' +jsonData.error.message + '</p>';
+  if (!input.trim()) {
+    showNotification('error', 'Please enter text to convert to speech');
     return;
   }
 
-  var div = document.getElementById('result');  // Get the div by its ID
-  var link=document.createElement('a');
-  link.className = "m-2 float-right inline-block rounded bg-primary px-6 pb-2.5 mb-3 pt-2.5 text-xs font-medium uppercase leading-normal text-white shadow-primary-3 transition duration-150 ease-in-out hover:bg-primary-accent-300 hover:shadow-primary-2 focus:bg-primary-accent-300 focus:shadow-primary-2 focus:outline-none focus:ring-0 active:bg-primary-600 active:shadow-primary-2 dark:shadow-black/30 dark:hover:shadow-dark-strong dark:focus:shadow-dark-strong dark:active:shadow-dark-strong";
-  link.innerHTML = "<i class='fa-solid fa-download'></i> Download result";
-  const blob = await response.blob();
-  link.href=window.URL.createObjectURL(blob);
+  tts(input);
+}
 
-  div.innerHTML = '';                             // Clear the existing content of the div
-  div.appendChild(link);                           // Add the new img element to the div
-  console.log(link)
-  document.getElementById("loader").style.display = "none";
-  document.getElementById("input").disabled = false;
+function showNotification(type, message) {
+  // Remove any existing notification
+  const existingNotification = document.getElementById('notification');
+  if (existingNotification) {
+    existingNotification.remove();
+  }
+  
+  // Create new notification
+  const notification = document.createElement('div');
+  notification.id = 'notification';
+  notification.classList.add(
+    'fixed', 'top-24', 'right-4', 'z-50', 'p-4', 'rounded-lg', 'shadow-lg',
+    'transform', 'transition-all', 'duration-300', 'ease-in-out', 'translate-y-0',
+    'flex', 'items-center', 'gap-2'
+  );
+  
+  // Style based on notification type
+  if (type === 'error') {
+    notification.classList.add('bg-red-900/90', 'border', 'border-red-700', 'text-red-200');
+    notification.innerHTML = '<i class="fas fa-circle-exclamation text-red-400 mr-2"></i>' + message;
+  } else if (type === 'warning') {
+    notification.classList.add('bg-yellow-900/90', 'border', 'border-yellow-700', 'text-yellow-200');
+    notification.innerHTML = '<i class="fas fa-triangle-exclamation text-yellow-400 mr-2"></i>' + message;
+  } else if (type === 'success') {
+    notification.classList.add('bg-green-900/90', 'border', 'border-green-700', 'text-green-200');
+    notification.innerHTML = '<i class="fas fa-circle-check text-green-400 mr-2"></i>' + message;
+  } else {
+    notification.classList.add('bg-blue-900/90', 'border', 'border-blue-700', 'text-blue-200');
+    notification.innerHTML = '<i class="fas fa-circle-info text-blue-400 mr-2"></i>' + message;
+  }
+  
+  // Add close button
+  const closeBtn = document.createElement('button');
+  closeBtn.innerHTML = '<i class="fas fa-xmark"></i>';
+  closeBtn.classList.add('ml-auto', 'text-gray-400', 'hover:text-white', 'transition-colors');
+  closeBtn.onclick = () => {
+    notification.classList.add('opacity-0', 'translate-y-[-20px]');
+    setTimeout(() => notification.remove(), 300);
+  };
+  notification.appendChild(closeBtn);
+  
+  // Add to DOM
+  document.body.appendChild(notification);
+  
+  // Animate in
+  setTimeout(() => {
+    notification.classList.add('opacity-0', 'translate-y-[-20px]');
+    notification.offsetHeight; // Force reflow
+    notification.classList.remove('opacity-0', 'translate-y-[-20px]');
+  }, 10);
+  
+  // Auto dismiss after 5 seconds
+  setTimeout(() => {
+    if (document.getElementById('notification')) {
+      notification.classList.add('opacity-0', 'translate-y-[-20px]');
+      setTimeout(() => notification.remove(), 300);
+    }
+  }, 5000);
+}
+
+async function tts(input) {
+  // Show loader and prepare UI
+  const loader = document.getElementById("loader");
+  const inputField = document.getElementById("input");
+  const resultDiv = document.getElementById("result");
+  
+  loader.style.display = "block";
+  inputField.value = "";
+  inputField.disabled = true;
+  resultDiv.innerHTML = '<div class="text-center text-gray-400 italic">Processing your request...</div>';
+
+  // Get the model and make API request
+  const model = document.getElementById("tts-model").value;
+  try {
+    const response = await fetch("tts", {
+      method: "POST",
+      headers: {
+        "Content-Type": "application/json",
+      },
+      body: JSON.stringify({
+        model: model,
+        input: input,
+      }),
+    });
+    
+    if (!response.ok) {
+      const jsonData = await response.json();
+      resultDiv.innerHTML = `
+        <div class="bg-red-900/30 border border-red-700/50 rounded-lg p-4 text-center">
+          <i class="fas fa-circle-exclamation text-red-400 text-2xl mb-2"></i>
+          <p class="text-red-300 font-medium">${jsonData.error.message || 'An error occurred'}</p>
+        </div>
+      `;
+      showNotification('error', 'Failed to generate audio');
+      return;
+    }
+
+    // Handle successful response
+    const blob = await response.blob();
+    const audioUrl = window.URL.createObjectURL(blob);
+    
+    // Create audio player
+    const audioPlayer = document.createElement('div');
+    audioPlayer.className = 'flex flex-col items-center space-y-4 w-full';
+    
+    // Create audio element with styled controls
+    const audio = document.createElement('audio');
+    audio.controls = true;
+    audio.src = audioUrl;
+    audio.className = 'w-full my-4';
+    audioPlayer.appendChild(audio);
+    
+    // Create action buttons container
+    const actionButtons = document.createElement('div');
+    actionButtons.className = 'flex flex-wrap justify-center gap-3';
+    
+    // Download button
+    const downloadLink = document.createElement('a');
+    downloadLink.href = audioUrl;
+    downloadLink.download = `tts-${model}-${new Date().toISOString().slice(0, 10)}.mp3`;
+    downloadLink.className = 'group flex items-center bg-blue-600 hover:bg-blue-700 text-white py-2 px-4 rounded-lg transition duration-300 ease-in-out transform hover:scale-105 hover:shadow-lg';
+    downloadLink.innerHTML = `
+      <i class="fas fa-download mr-2"></i>
+      <span>Download</span>
+      <i class="fas fa-arrow-right opacity-0 group-hover:opacity-100 group-hover:translate-x-2 ml-2 transition-all duration-300"></i>
+    `;
+    actionButtons.appendChild(downloadLink);
+    
+    // Replay button
+    const replayButton = document.createElement('button');
+    replayButton.className = 'group flex items-center bg-purple-600 hover:bg-purple-700 text-white py-2 px-4 rounded-lg transition duration-300 ease-in-out transform hover:scale-105 hover:shadow-lg';
+    replayButton.innerHTML = `
+      <i class="fas fa-rotate-right mr-2"></i>
+      <span>Replay</span>
+    `;
+    replayButton.onclick = () => audio.play();
+    actionButtons.appendChild(replayButton);
+    
+    // Add text display
+    const textDisplay = document.createElement('div');
+    textDisplay.className = 'mt-4 p-4 bg-gray-800/50 border border-gray-700/50 rounded-lg text-gray-300 text-center italic';
+    textDisplay.textContent = `"${input}"`;
+    
+    // Add all elements to result div
+    audioPlayer.appendChild(actionButtons);
+    resultDiv.innerHTML = '';
+    resultDiv.appendChild(audioPlayer);
+    resultDiv.appendChild(textDisplay);
+    
+    // Play audio automatically
+    audio.play();
+    
+    // Show success notification
+    showNotification('success', 'Audio generated successfully');
+    
+  } catch (error) {
+    console.error('Error generating audio:', error);
+    resultDiv.innerHTML = `
+      <div class="bg-red-900/30 border border-red-700/50 rounded-lg p-4 text-center">
+        <i class="fas fa-circle-exclamation text-red-400 text-2xl mb-2"></i>
+        <p class="text-red-300 font-medium">Network error: Failed to connect to the server</p>
+      </div>
+    `;
+    showNotification('error', 'Network error occurred');
+  } finally {
+    // Reset UI state
+    loader.style.display = "none";
+    inputField.disabled = false;
+    inputField.focus();
+  }
+}
+
+// Set up event listeners when DOM is loaded
+document.addEventListener('DOMContentLoaded', () => {
   document.getElementById("input").focus();
-}
-
-document.getElementById("key").addEventListener("submit", submitKey);
-document.getElementById("input").focus();
-document.getElementById("tts").addEventListener("submit", genAudio);
-document.getElementById("loader").style.display = "none";
-
-const storeKey = localStorage.getItem("key");
-if (storeKey) {
-  document.getElementById("apiKey").value = storeKey;
-}
-
+  document.getElementById("tts").addEventListener("submit", genAudio);
+  document.getElementById("loader").style.display = "none";
+ 
+  // Add basic keyboard shortcuts
+  document.addEventListener('keydown', (e) => {
+    // Submit on Ctrl+Enter
+    if (e.key === 'Enter' && e.ctrlKey && document.activeElement.id === 'input') {
+      e.preventDefault();
+      document.getElementById("tts").dispatchEvent(new Event('submit'));
+    }
+  });
+});

core/http/views/404.html

@@ -1,28 +1,51 @@
 <!DOCTYPE html>
 <html lang="en">
-
 {{template "views/partials/head" .}}
 
-<body class="bg-black text-white">
+<body class="bg-gradient-to-br from-gray-900 to-gray-950 text-gray-200">
 <div class="flex flex-col min-h-screen">
-   
+
     {{template "views/partials/navbar" .}}
-    
-    <div class="container mx-auto px-4 flex-grow">
-        <div class="header text-center py-12">
-            <h1 class="text-5xl font-bold">Welcome to your LocalAI instance!</h1>
-            <div class="mt-6">
-         <!--       <a href="./" aria-label="HomePage" alt="HomePage">
-                    <img class="mx-auto w-1/4 h-auto" src="https://github.com/go-skynet/LocalAI/assets/2420543/0966aa2a-166e-4f99-a3e5-6c915fc997dd" alt="LocalAI Logo">            
-                </a>
-            -->
+
+    <div class="container mx-auto px-4 py-8 flex-grow">
+        <!-- Error Section -->
+        <div class="bg-gradient-to-r from-blue-900/30 to-indigo-900/30 rounded-2xl shadow-xl p-8 mb-10">
+            <div class="max-w-4xl mx-auto text-center">
+                <div class="mb-6 text-6xl text-blue-400">
+                    <i class="fas fa-exclamation-circle"></i>
+                </div>
+                <h1 class="text-4xl md:text-5xl font-bold text-white mb-4">
+                    <span class="bg-clip-text text-transparent bg-gradient-to-r from-blue-400 to-indigo-400">
+                        404 - Page Not Found
+                    </span>
+                </h1>
+                <p class="text-xl text-gray-300 mb-6">The page you're looking for doesn't exist or has been moved</p>
+                <div class="flex flex-wrap justify-center gap-4">
+                    <a href="./" 
+                       class="group flex items-center bg-blue-600 hover:bg-blue-700 text-white py-2 px-6 rounded-lg transition duration-300 ease-in-out transform hover:scale-105 hover:shadow-lg">
+                        <i class="fas fa-home mr-2"></i>
+                        <span>Return Home</span>
+                        <i class="fas fa-arrow-right opacity-0 group-hover:opacity-100 group-hover:translate-x-2 ml-2 transition-all duration-300"></i>
+                    </a>
+                    <a href="browse" 
+                       class="group flex items-center bg-indigo-600 hover:bg-indigo-700 text-white py-2 px-6 rounded-lg transition duration-300 ease-in-out transform hover:scale-105 hover:shadow-lg">
+                        <i class="fas fa-images mr-2"></i>
+                        <span>Browse Gallery</span>
+                        <i class="fas fa-arrow-right opacity-0 group-hover:opacity-100 group-hover:translate-x-2 ml-2 transition-all duration-300"></i>
+                    </a>
+                </div>
             </div>
-            <p class="mt-4 text-lg">The FOSS alternative to OpenAI, Claude, ...</p>
-            <a href="https://localai.io" target="_blank" class="mt-4 inline-block bg-blue-500 text-white py-2 px-4 rounded transition duration-300 ease-in-out hover:bg-blue-700"><i class="fas fa-book-reader pr-2"></i>Documentation</a>
         </div>
 
-        <div class="models mt-12">
-            <h2 class="text-center text-3xl font-semibold">Nothing found!</h2>
+        <!-- Additional Information -->
+        <div class="bg-gray-800/50 border border-gray-700/50 rounded-xl p-8 shadow-md backdrop-blur-sm">
+            <div class="text-center max-w-3xl mx-auto">
+                <div class="inline-flex items-center justify-center w-16 h-16 rounded-full bg-yellow-500/20 mb-4">
+                    <i class="text-yellow-400 text-2xl fa-solid fa-triangle-exclamation"></i>
+                </div>
+                <h2 class="text-2xl md:text-3xl font-semibold text-gray-100 mb-4">Looking for resources?</h2>
+                <p class="text-lg text-gray-300 mb-6">Visit our <a class="text-blue-400 hover:text-blue-300 underline underline-offset-2" href="browse">🖼️ Gallery</a> or check the <a href="https://localai.io/basics/getting_started/" class="text-blue-400 hover:text-blue-300 underline underline-offset-2"> <i class="fa-solid fa-book"></i> Getting started documentation</a></p>
+            </div>
         </div>
     </div>
 
@@ -30,4 +53,4 @@
 </div>
 
 </body>
-</html>
+</html>

core/http/views/chat.html

@@ -4,7 +4,7 @@ Part of this page is based on the OpenAI Chatbot example by David Härer:
 https://github.com/david-haerer/chatapi
 
 MIT License Copyright (c) 2023 David Härer
-            Copyright (c) 2024 Ettore Di Giacinto
+            Copyright (c) 2024-2025 Ettore Di Giacinto
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal
@@ -29,145 +29,355 @@ SOFTWARE.
 <html lang="en">
   {{template "views/partials/head" .}}
   <script defer src="static/chat.js"></script>
-  <style>
-    body {
-        overflow: hidden; 
-    }
-  </style>
-  <body class="bg-gray-900 text-gray-200" x-data="{ key: $store.chat.key }">
-    <div class="flex flex-col min-h-screen">
-
+  {{ $allGalleryConfigs:=.GalleryConfig }}
+  {{ $model:=.Model}}
+  <body class="bg-slate-900 text-gray-100 flex flex-col h-screen" x-data="{ sidebarOpen: true }">
     {{template "views/partials/navbar" .}}
-    <div class="chat-container mt-2 mr-2 ml-2 mb-2 bg-gray-800 shadow-lg rounded-lg" >
-     <!-- Chat Header -->
-    <div class="border-b border-gray-700 p-4"  x-data="{ component: 'menu' }">
 
-      <div class="flex items-center justify-between">
+    <!-- Main container with sidebar toggle -->
+    <div class="flex flex-1 overflow-hidden relative">
+      <!-- Sidebar -->
+      <div 
+        class="sidebar bg-gray-800 fixed top-16 bottom-0 left-0 w-64 transform transition-transform duration-300 ease-in-out z-30 border-r border-gray-700 overflow-y-auto"
+        :class="sidebarOpen ? 'translate-x-0' : '-translate-x-full'">
+        
+        <div class="p-4 flex justify-between items-center border-b border-gray-700">
+          <h2 class="text-lg font-semibold">Chat Settings</h2>
+          <button 
+            @click="sidebarOpen = false"
+            class="text-gray-400 hover:text-white focus:outline-none">
+            <i class="fa-solid fa-times"></i>
+          </button>
+        </div>
 
-      <h1 class="text-lg font-semibold"> <i class="fa-solid fa-comments"></i> Chat with {{.Model}} <a href="https://localai.io/features/text-generation/" target="_blank" >
-        <i class="fas fa-circle-info pr-2"></i>
-      </a></h1>
-      <div x-show="component === 'menu'" id="menu">
-        <button
-          @click="$store.chat.clear()"
-          id="clear"
-          title="Clear chat history"
+        <!-- Sidebar content -->
+        <div class="p-4 space-y-6">
+          <!-- Model selection - Fixed to properly select current model -->
+          <div class="space-y-2">
+            <label class="text-sm font-medium text-gray-300">Select Model</label>
+            <select
+              id="modelSelector"
+              class="w-full bg-gray-700 text-white border border-gray-600 focus:border-blue-500 focus:ring focus:ring-blue-500 focus:ring-opacity-50 rounded-md shadow-sm p-2 appearance-none"
+              onchange="window.location = this.value"
+            >
+              <option value="" disabled class="text-gray-400">Select a model</option>
+             
+              {{ range .ModelsConfig }}
+                {{ $cfg := . }}
+                {{ range .KnownUsecaseStrings }}
+                  {{ if eq . "FLAG_CHAT" }}
+                    <option
+                      value="chat/{{$cfg.Name}}"
+                      {{ if eq $cfg.Name $model }} selected {{end}}
+                      class="bg-gray-700 text-white"
+                    >
+                      {{$cfg.Name}}
+                    </option>
+                  {{ end }}
+                {{ end }}
+              {{ end }}
+              {{ range .ModelsWithoutConfig }}
+                <option
+                  value="chat/{{.}}"
+                  {{ if eq . $model }} selected {{ end }}
+                  class="bg-gray-700 text-white"
+                >
+                  {{.}}
+                </option>
+              {{end}}
+            </select>
+          </div>
 
-          data-twe-ripple-init
-          data-twe-ripple-color="light"
-          class="m-2 float-right inline-block rounded bg-primary px-6 pb-2.5 mb-3 pt-2.5 text-xs font-medium uppercase leading-normal text-white shadow-primary-3 transition duration-150 ease-in-out hover:bg-primary-accent-300 hover:shadow-primary-2 focus:bg-primary-accent-300 focus:shadow-primary-2 focus:outline-none focus:ring-0 active:bg-primary-600 active:shadow-primary-2 dark:shadow-black/30 dark:hover:shadow-dark-strong dark:focus:shadow-dark-strong dark:active:shadow-dark-strong"
-          >
-          Clear chat 🔥
-        </button>
-        <button @click="component = 'key'" title="Update API key"
-        class="m-2 float-right inline-block rounded bg-primary px-6 pb-2.5 mb-3 pt-2.5 text-xs font-medium uppercase leading-normal text-white shadow-primary-3 transition duration-150 ease-in-out hover:bg-primary-accent-300 hover:shadow-primary-2 focus:bg-primary-accent-300 focus:shadow-primary-2 focus:outline-none focus:ring-0 active:bg-primary-600 active:shadow-primary-2 dark:shadow-black/30 dark:hover:shadow-dark-strong dark:focus:shadow-dark-strong dark:active:shadow-dark-strong"
-        >Set API Key🔑</button>
-        <button @click="component = 'system_prompt'" title="System Prompt"
-        class="m-2 float-right inline-block rounded bg-primary px-6 pb-2.5 mb-3 pt-2.5 text-xs font-medium uppercase leading-normal text-white shadow-primary-3 transition duration-150 ease-in-out hover:bg-primary-accent-300 hover:shadow-primary-2 focus:bg-primary-accent-300 focus:shadow-primary-2 focus:outline-none focus:ring-0 active:bg-primary-600 active:shadow-primary-2 dark:shadow-black/30 dark:hover:shadow-dark-strong dark:focus:shadow-dark-strong dark:active:shadow-dark-strong"
-        >Set system prompt</button>
+          {{ if $model }}
+          {{ $galleryConfig:= index $allGalleryConfigs $model}}
+          {{ if $galleryConfig }}
+          <!-- Model info -->
+          <div class="space-y-2">
+            <div class="flex items-center">
+              {{ if $galleryConfig.Icon }}<img src="{{$galleryConfig.Icon}}" class="rounded-lg w-8 h-8 mr-2">{{end}}
+              <h3 class="text-md font-medium">{{ $model }}</h3>
+            </div>
+            <button data-twe-ripple-init data-twe-ripple-color="light" class="w-full text-left flex items-center px-3 py-2 text-xs rounded text-white bg-gray-700 hover:bg-gray-600 transition-colors" data-modal-target="model-info-modal" data-modal-toggle="model-info-modal">
+              <i class="fas fa-info-circle mr-2"></i>
+              Model Information
+            </button>
+          </div>
+          {{ end }}
+          {{ end }}
+
+          <div x-data="{ activeTab: 'actions' }" class="space-y-4">
+            <!-- Tab navigation -->
+            <div class="flex border-b border-gray-700">
+              <button 
+                @click="activeTab = 'actions'" 
+                :class="activeTab === 'actions' ? 'border-b-2 border-blue-500 text-white' : 'text-gray-400 hover:text-white'"
+                class="py-2 px-4 text-sm font-medium">
+                Actions
+              </button>
+              <button 
+                @click="activeTab = 'settings'" 
+                :class="activeTab === 'settings' ? 'border-b-2 border-blue-500 text-white' : 'text-gray-400 hover:text-white'"
+                class="py-2 px-4 text-sm font-medium">
+                Settings
+              </button>
+            </div>
+
+            <!-- Actions tab -->
+            <div x-show="activeTab === 'actions'" class="space-y-3">
+              <button
+                @click="$store.chat.clear()"
+                id="clear"
+                title="Clear chat history"
+                class="w-full flex items-center px-3 py-2 text-sm rounded text-white bg-gray-700 hover:bg-gray-600 transition-colors"
+              >
+                <i class="fa-solid fa-trash-can mr-2"></i> Clear chat
+              </button>
+              
+              <a 
+                href="https://localai.io/features/text-generation/" 
+                target="_blank"
+                class="w-full flex items-center px-3 py-2 text-sm rounded text-white bg-gray-700 hover:bg-gray-600 transition-colors"
+              >
+                <i class="fas fa-book mr-2"></i> Documentation
+              </a>
+              
+              <a 
+                href="browse?term={{.Model}}" 
+                class="w-full flex items-center px-3 py-2 text-sm rounded text-white bg-gray-700 hover:bg-gray-600 transition-colors"
+              >
+                <i class="fas fa-brain mr-2"></i> Browse Model
+              </a>
+            </div>
+
+            <!-- Settings tab -->
+            <div x-show="activeTab === 'settings'" x-data="{ showPromptForm: false }" class="space-y-3">           
+              <button 
+                @click="showPromptForm = !showPromptForm" 
+                class="w-full flex items-center justify-between px-3 py-2 text-sm rounded text-white bg-gray-700 hover:bg-gray-600 transition-colors"
+              >
+                <span><i class="fa-solid fa-message mr-2"></i> System Prompt</span>
+                <i :class="showPromptForm ? 'fa-chevron-up' : 'fa-chevron-down'" class="fa-solid"></i>
+              </button>
+              
+              <div x-show="showPromptForm" class="p-3 bg-gray-700 rounded">
+                <form id="system_prompt" class="flex flex-col space-y-2">
+                  <textarea
+                    type="text"
+                    id="systemPrompt"
+                    name="systemPrompt"
+                    class="bg-gray-800 text-white border border-gray-600 focus:border-blue-500 focus:ring focus:ring-blue-500 focus:ring-opacity-50 rounded-md shadow-sm p-2 appearance-none min-h-24"
+                    placeholder="System prompt"
+                    x-model.lazy="$store.chat.systemPrompt"
+                  ></textarea>
+                  <button
+                    type="submit"
+                    class="px-3 py-2 text-sm rounded text-white bg-blue-600 hover:bg-blue-700 transition-colors"
+                  >
+                    Save System Prompt
+                  </button>
+                </form>
+              </div>
+            </div>
+          </div>
+        </div>
       </div>
-      <form x-show="component === 'key'" id="key">
-        <input
-          type="password"
-          id="apiKey"
-          name="apiKey"
-          class="bg-gray-800 text-white border border-gray-600 focus:border-blue-500 focus:ring focus:ring-blue-500 focus:ring-opacity-50 rounded-md shadow-sm p-2 appearance-none"
-          placeholder="OpenAI API Key"
-          x-model.lazy="key"
-        />
-        <button @click="component = 'menu'" type="submit" title="Save API key">
-          <i class="fa-solid fa-arrow-right"></i>
-        </button>
-      </form>
-      <form x-show="component === 'system_prompt'" id="system_prompt">
-        <textarea
-          type="text"
-          id="systemPrompt"
-          name="systemPrompt"
-          class="bg-gray-800 text-white border border-gray-600 focus:border-blue-500 focus:ring focus:ring-blue-500 focus:ring-opacity-50 rounded-md shadow-sm p-2 appearance-none"
-          placeholder="System prompt"
-          x-model.lazy="system_prompt"
-        ></textarea>
-        <button @click="component = 'menu'" type="submit" title="Save Prompt">
-          <i class="fa-solid fa-arrow-right"></i>
-        </button>
-      </form>
 
-      <select x-data="{ link : '' }" x-model="link" x-init="$watch('link', value => window.location = link)" 
-        class="bg-gray-800 text-white border border-gray-600 focus:border-blue-500 focus:ring focus:ring-blue-500 focus:ring-opacity-50 rounded-md shadow-sm p-2 appearance-none"
-        >	
-        <!-- Options -->
-        <option value="" disabled class="text-gray-400" >Select a model</option>
-        {{ $model:=.Model}}
-        {{ range .ModelsConfig }}
-        {{ if eq . $model }}
-        <option value="chat/{{.}}" selected  class="bg-gray-700 text-white">{{.}}</option>
-        {{ else }}
-        <option value="chat/{{.}}" class="bg-gray-700 text-white">{{.}}</option>
-        {{ end }}
-        {{ end }}
-      </select>
+      <!-- Main chat container (shifts with sidebar) -->
+      <div 
+        class="flex-1 flex flex-col transition-all duration-300 ease-in-out"
+        :class="sidebarOpen ? 'ml-64' : 'ml-0'">
+        
+        <!-- Chat header with toggle button -->
+        <div class="border-b border-gray-700 p-4 flex items-center">
+          <!-- Sidebar toggle button moved to be the first element in the header and with clear styling -->
+          <button 
+            @click="sidebarOpen = !sidebarOpen" 
+            class="mr-4 text-gray-300 hover:text-white focus:outline-none bg-gray-800 hover:bg-gray-700 p-2 rounded"
+            style="min-width: 36px;"
+            title="Toggle settings">
+            <i class="fa-solid" :class="sidebarOpen ? 'fa-times' : 'fa-bars'"></i>
+          </button>
+          
+          <div class="flex items-center">
+            <i class="fa-solid fa-comments mr-2"></i>
+            {{ if $model }}
+            {{ $galleryConfig:= index $allGalleryConfigs $model}}
+            {{ if $galleryConfig }}
+            {{ if $galleryConfig.Icon }}<img src="{{$galleryConfig.Icon}}" class="rounded-lg w-8 h-8 mr-2">{{end}}
+            {{ end }}
+            {{ end }}
+            <h1 class="text-lg font-semibold">
+              Chat {{ if .Model }} with {{.Model}} {{ end }}
+            </h1>
+          </div>
+        </div>
 
-      </div>
-    </div>
-
-    <div class="chat-messages p-4" id="chat" x-data="{history: $store.chat.history}">
-      <p id="usage" x-show="history.length === 0">
-        Start chatting with the AI by typing a prompt in the input field below and pressing Enter.
-        For models that support images, you can upload an image by clicking the paperclip <i class="fa-solid fa-paperclip"></i> icon.
-      </p>
-      <div id="messages">
-      <template x-for="message in history">
-        <div class="message flex items-start space-x-2 my-2" >
-          <!--<img :src="message.role === 'user' ? '/path/to/user-icon.png' : '/path/to/bot-icon.png'" alt="" class="h-6 w-6">-->
-          <i class="fa-solid h-8 w-8" :class="message.role === 'user' ? 'fa-user' : 'fa-robot'"  ></i>
-          <div class="flex flex-col flex-1">
-            <span class="text-xs font-semibold text-gray-600" x-text="message.role === 'user' ? 'User' : 'Assistant ({{.Model}})'"></span>
-            <template x-if="message.role === 'user'">
-              <div class="p-2 flex-1 rounded" :class="message.role" x-html="message.html"></div>
-            </template>
-            <template x-if="message.role === 'assistant'">
-              <div class="p-2 flex-1 rounded" :class="message.role" x-html="message.html"></div>
-            </template>
-            <template x-if="message.image">
-              <img :src="message.image" alt="Image" class="rounded-lg mt-2 h-36 w-36">
+        <!-- Chat messages area -->
+        <div class="flex-1 p-4 overflow-auto" id="chat" x-data="{history: $store.chat.history}">
+          <p id="usage" x-show="history.length === 0" class="text-gray-300">
+            Start chatting with the AI by typing a prompt in the input field below and pressing Enter.
+            For models that support images, you can upload an image by clicking the paperclip
+            <i class="fa-solid fa-paperclip"></i> icon.
+          </p>
+          <div id="messages" class="max-w-3xl mx-auto">
+            <template x-for="message in history">
+              <div :class="message.role === 'user' ? 'flex items-start space-x-2 my-2 justify-end' : 'flex items-start space-x-2 my-2'">
+                {{ if .Model }}
+                {{ $galleryConfig:= index $allGalleryConfigs .Model}}
+                <template x-if="message.role === 'user'">
+                  <div class="flex items-center space-x-2">
+                    <div class="flex flex-col flex-1 items-end">
+                      <span class="text-xs font-semibold text-gray-400">You</span>
+                      <div class="p-2 flex-1 rounded bg-gray-700 text-white" x-html="message.html"></div>
+                      <template x-if="message.image">
+                        <img :src="message.image" alt="Image" class="rounded-lg mt-2 max-w-xs">
+                      </template>
+                    </div>
+                  </div>
+                </template>
+                <template x-if="message.role != 'user'">
+                  <div class="flex items-center space-x-2">
+                    {{ if $galleryConfig }}
+                    {{ if $galleryConfig.Icon }}<img src="{{$galleryConfig.Icon}}" class="rounded-lg mt-2 max-w-8 max-h-8">{{end}}
+                    {{ end }}
+                    <div class="flex flex-col flex-1">
+                      <span class="text-xs font-semibold text-gray-400">{{if .Model}}{{.Model}}{{else}}Assistant{{end}}</span>
+                      <div class="flex-1 text-white flex items-center space-x-2">
+                        <div x-html="message.html"></div>
+                        <button @click="copyToClipboard(message.html)" title="Copy to clipboard" class="text-gray-400 hover:text-gray-100">
+                          <i class="fa-solid fa-copy"></i>
+                        </button>
+                      </div>
+                      <template x-if="message.image">
+                        <img :src="message.image" alt="Image" class="rounded-lg mt-2 max-w-xs">
+                      </template>
+                    </div>
+                  </div>
+                </template>
+                {{ else }}
+                <i
+                  class="fa-solid h-8 w-8"
+                  :class="message.role === 'user' ? 'fa-user' : 'fa-robot'"
+                ></i>
+                {{ end }}
+              </div>
             </template>
           </div>
         </div>
-      </template>
-      </div>
-    </div>
 
-    <div class="p-4 border-t border-gray-700" x-data="{ inputValue: '', shiftPressed: false, fileName: ''  }">
-      <div id="loader" class="my-2 loader" style="display: none;"></div>
-      <input id="chat-model" type="hidden" value="{{.Model}}">
-      <input id="input_image" type="file" style="display: none;" @change="fileName = $event.target.files[0].name">
-      <form id="prompt" action="chat/{{.Model}}" method="get" @submit.prevent="submitPrompt">
-          <div class="relative w-full">
-              <textarea
+
+          <!-- Chat Input -->
+          <div class="p-4 border-t border-gray-700" x-data="{ inputValue: '', shiftPressed: false, fileName: '', isLoading: false }">
+            <form id="prompt" action="chat/{{.Model}}" method="get" @submit.prevent="submitPrompt" class="max-w-3xl mx-auto">
+              <div class="relative w-full bg-gray-800 rounded-xl shadow-md">
+                <textarea
                   id="input"
                   name="input"
                   x-model="inputValue"
                   placeholder="Send a message..."
-                  class="p-2 pl-2 border rounded w-full bg-gray-600 text-white placeholder-gray-300"
+                  class="p-4 pr-16 w-full bg-gray-800 text-gray-100 placeholder-gray-400 focus:outline-none resize-none border-0 rounded-xl transition-colors duration-200"
                   required
                   @keydown.shift="shiftPressed = true"
                   @keyup.shift="shiftPressed = false"
                   @keydown.enter="if (!shiftPressed) { submitPrompt($event); }"
-                  style="padding-right: 4rem;"
-              ></textarea>
-              <span x-text="fileName" id="fileName" class="absolute right-16 top-5 text-gray-300 text-sm mr-2"></span>
-              <button type="button" onclick="document.getElementById('input_image').click()" class="fa-solid fa-paperclip text-gray-300 ml-2 absolute right-10 top-3 text-lg p-2">
-              </button>
-              <button type=submit><i class="fa-solid fa-circle-up text-gray-300 absolute right-2 top-3 text-lg p-2"></i></button>
+                  rows="3"
+                  style="box-shadow: 0 0 0 1px rgba(75, 85, 99, 0.4) inset;"
+                ></textarea>
+                <span x-text="fileName" id="fileName" class="absolute right-16 top-4 text-gray-400 text-sm mr-2"></span>
+                <button
+                  type="button"
+                  onclick="document.getElementById('input_image').click()"
+                  class="fa-solid fa-paperclip text-gray-400 absolute right-12 top-4 text-lg p-2 hover:text-blue-400 transition-colors duration-200"
+                  title="Attach an image"
+                ></button>
+                
+                <!-- Send button and loader in the same position -->
+                <div class="absolute right-3 top-4">
+                  <!-- Loader (hidden by default) -->
+                  <div id="loader" class="text-lg p-2" style="display: none;">
+                    <svg class="animate-spin h-5 w-5 text-blue-500" xmlns="http://www.w3.org/2000/svg" fill="none" viewBox="0 0 24 24">
+                      <circle class="opacity-25" cx="12" cy="12" r="10" stroke="currentColor" stroke-width="4"></circle>
+                      <path class="opacity-75" fill="currentColor" d="M4 12a8 8 0 018-8V0C5.373 0 0 5.373 0 12h4zm2 5.291A7.962 7.962 0 014 12H0c0 3.042 1.135 5.824 3 7.938l3-2.647z"></path>
+                    </svg>
+                  </div>
+                  
+                  <!-- Send button -->
+                  <button
+                    id="send-button"
+                    type="submit"
+                    class="text-lg p-2 text-gray-400 hover:text-blue-400 transition-colors duration-200"
+                    title="Send message"
+                  >
+                    <i class="fa-solid fa-paper-plane"></i>
+                  </button>
+                </div>
+              </div>
+            </form>
+            <input id="chat-model" type="hidden" value="{{.Model}}">
+            <input
+              id="input_image"
+              type="file"
+              style="display: none;"
+              @change="fileName = $event.target.files[0].name"
+            />
           </div>
-      </form>
-  </div>
+          </form>
+        </div>
+      </div>
+    </div>
+    
+    <!-- Modal moved outside of sidebar to appear in center of page -->
+    {{ if $model }}
+    {{ $galleryConfig:= index $allGalleryConfigs $model}}
+    {{ if $galleryConfig }}
+    <div id="model-info-modal" tabindex="-1" aria-hidden="true" class="hidden overflow-y-auto overflow-x-hidden fixed top-0 right-0 left-0 z-50 flex justify-center items-center w-full md:inset-0 h-[calc(100%-1rem)] max-h-full">
+      <div class="relative p-4 w-full max-w-2xl max-h-full">
+        <div class="relative p-4 w-full max-w-2xl max-h-full bg-white rounded-lg shadow dark:bg-gray-700">
+          <!-- Header -->
+          <div class="flex items-center justify-between p-4 md:p-5 border-b rounded-t dark:border-gray-600">
+            <h3 class="text-xl font-semibold text-gray-900 dark:text-white">{{ $model }}</h3>
+            <button class="text-gray-400 bg-transparent hover:bg-gray-200 hover:text-gray-900 rounded-lg text-sm w-8 h-8 ms-auto inline-flex justify-center items-center dark:hover:bg-gray-600 dark:hover:text-white" data-modal-hide="model-info-modal">
+              <svg class="w-3 h-3" aria-hidden="true" xmlns="http://www.w3.org/2000/svg" fill="none" viewBox="0 0 14 14">
+                <path stroke="currentColor" stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="m1 1 6 6m0 0 6 6M7 7l6-6M7 7l-6 6"/>
+              </svg>
+              <span class="sr-only">Close modal</span>
+            </button>
+          </div>
+                  
+          <!-- Body -->
+          <div class="p-4 md:p-5 space-y-4">
+            <div class="flex justify-center items-center">
+              {{ if $galleryConfig.Icon }}<img class="lazy rounded-t-lg max-h-48 max-w-96 object-cover mt-3 entered loaded" src="{{$galleryConfig.Icon}}" loading="lazy"/>{{end}}
+            </div>
+            <p class="text-base leading-relaxed text-gray-500 dark:text-gray-400">{{ $galleryConfig.Description }}</p>
+            <hr>
+            <p class="text-sm font-semibold text-gray-900 dark:text-white">Links</p>
+            <ul>
+              {{range $galleryConfig.URLs}}
+              <li><a href="{{ . }}" target="_blank">{{ . }}</a></li>
+              {{end}}
+            </ul>                   
+          </div>
+                  
+          <!-- Footer -->
+          <div class="flex items-center p-4 md:p-5 border-t border-gray-200 rounded-b dark:border-gray-600">
+            <button data-modal-hide="model-info-modal" class="py-2.5 px-5 ms-3 text-sm font-medium text-gray-900 focus:outline-none bg-white rounded-lg border border-gray-200 hover:bg-gray-100 hover:text-blue-700 focus:z-10 focus:ring-4 focus:ring-gray-100 dark:focus:ring-gray-700 dark:bg-gray-800 dark:text-gray-400 dark:border-gray-600 dark:hover:text-white dark:hover:bg-gray-700">
+              Close
+            </button>
+          </div>
+        </div>
+      </div>
+    </div>
+    {{ end }}
+    {{ end }}
+
+    <!-- Alpine store initialization -->
     <script>
       document.addEventListener("alpine:init", () => {
         Alpine.store("chat", {
           history: [],
           languages: [undefined],
+          systemPrompt: "",
           clear() {
             this.history.length = 0;
           },
@@ -175,31 +385,22 @@ SOFTWARE.
             const N = this.history.length - 1;
             if (this.history.length && this.history[N].role === role) {
               this.history[N].content += content;
-              str = this.history[N].content;
               this.history[N].html = DOMPurify.sanitize(
-                marked.parse(this.history[N].content),
+                marked.parse(this.history[N].content)
               );
             } else {
-              c =  ""
-              // split content newlines in content
+              let c = "";
               const lines = content.split("\n");
-              // for each line, do DOMPurify.sanitize(marked.parse(line)) and add it to c
               lines.forEach((line) => {
                 c += DOMPurify.sanitize(marked.parse(line));
               });
-
-              this.history.push({
-                role: role,
-                content: content,
-                html: c,
-                image: image,
-              });
+              this.history.push({ role, content, html: c, image });
             }
-
+            document.getElementById('messages').scrollIntoView(false);
             const parser = new DOMParser();
             const html = parser.parseFromString(
               this.history[this.history.length - 1].html,
-              "text/html",
+              "text/html"
             );
             const code = html.querySelectorAll("pre code");
             if (!code.length) return;
@@ -213,17 +414,26 @@ SOFTWARE.
             });
           },
           messages() {
-            return this.history.map((message) => {
-              return {
-                role: message.role,
-                content: message.content,
-                image: message.image,
-              };
-            });
+            return this.history.map((message) => ({
+              role: message.role,
+              content: message.content,
+              image: message.image,
+            }));
           },
         });
+
+        window.copyToClipboard = (content) => {
+          const tempElement = document.createElement('div');
+          tempElement.innerHTML = content;
+          const text = tempElement.textContent || tempElement.innerText;
+
+          navigator.clipboard.writeText(text).then(() => {
+            alert('Copied to clipboard!');
+          }).catch(err => {
+            console.error('Failed to copy: ', err);
+          });
+        };
       });
     </script>
-    </div>
   </body>
-</html>
+</html>

core/http/views/explorer.html

@@ -1,380 +1,224 @@
 <!DOCTYPE html>
 <html lang="en">
-
 {{template "views/partials/head" .}}
 
-<style>
-    body {
-        background-color: #1a202c;
-        color: #e2e8f0;
-        font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
-    }
-    .token {
-        word-break: break-all;
-    }
-    .container {
-        max-width: 800px;
-        margin: 0 auto;
-        padding: 20px;
-        position: relative;
-    }
-    .network-card {
-        background-color: #2d3748;
-        padding: 20px;
-        border-radius: 8px;
-        margin-bottom: 20px;
-        box-shadow: 0 4px 6px rgba(0, 0, 0, 0.1);
-        transition: transform 0.3s ease, box-shadow 0.3s ease;
-    }
-    .network-card:hover {
-        transform: translateY(-5px);
-        box-shadow: 0 6px 10px rgba(0, 0, 0, 0.15);
-    }
-    .network-title {
-        font-size: 24px;
-        font-weight: bold;
-        margin-bottom: 10px;
-        color: #63b3ed;
-    }
-    .network-token {
-        font-size: 14px;
-        font-style: italic;
-        color: #cbd5e0;
-        margin-bottom: 10px;
-        word-break: break-word; /* Breaks words to prevent overflow */
-        overflow-wrap: break-word; /* Ensures long strings break */
-        white-space: pre-wrap; /* Preserves whitespace for breaking */
-    }
-    .cluster {
-        margin-top: 10px;
-        background-color: #4a5568;
-        padding: 10px;
-        border-radius: 6px;
-        transition: background-color 0.3s ease;
-    }
-    .cluster:hover {
-        background-color: #5a6b78;
-    }
-    .cluster-title {
-        font-size: 18px;
-        font-weight: bold;
-        color: #e2e8f0;
-    }
-    .form-container {
-        background-color: #2d3748;
-        padding: 20px;
-        border-radius: 8px;
-        margin-bottom: 20px;
-        box-shadow: 0 4px 6px rgba(0, 0, 0, 0.1);
-    }
-    .form-control {
-        margin-bottom: 15px;
-    }
-    label {
-        display: block;
-        margin-bottom: 5px;
-        font-weight: bold;
-    }
-    input[type="text"],
-    textarea {
-        width: 100%;
-        padding: 10px;
-        border-radius: 4px;
-        border: 1px solid #4a5568;
-        background-color: #3a4250;
-        color: #e2e8f0;
-        transition: border-color 0.3s ease, background-color 0.3s ease;
-    }
-    input[type="text"]:focus,
-    textarea:focus {
-        border-color: #63b3ed;
-        background-color: #4a5568;
-    }
-    button {
-        background-color: #3182ce;
-        color: #e2e8f0;
-        padding: 10px 20px;
-        border: none;
-        border-radius: 4px;
-        cursor: pointer;
-        transition: background-color 0.3s ease;
-    }
-    .error {
-        color: #e53e3e;
-        margin-top: 5px;
-    }
-    .success {
-        color: #38a169;
-        margin-top: 5px;
-    }
-    /* Spinner Styles */
-    .spinner {
-        display: inline-block;
-        width: 50px;
-        height: 50px;
-        border: 5px solid rgba(255, 255, 255, 0.2);
-        border-radius: 50%;
-        border-top-color: #3182ce;
-        animation: spin 1s linear infinite;
-        margin: 0 auto;
-    }
+<body class="bg-gradient-to-br from-gray-900 to-gray-950 text-gray-200">
+<div class="flex flex-col min-h-screen">
 
-    @keyframes spin {
-        to { transform: rotate(360deg); }
-    }
-
-    /* Center the loading text and spinner */
-    .loading-container {
-        text-align: center;
-        padding: 50px;
-    }
-    .warning-box {
-            border-radius: 5px;
-    }
-    .warning-box i {
-        margin-right: 10px;
-    }
-    .token-box {
-            background-color: #4a5568;
-            padding: 10px;
-            border-radius: 4px;
-            margin-top: 10px;
-            position: relative;
-            cursor: pointer;
-        }
-        .token-box:hover {
-            background-color: #5a6b7e;
-        }
-        .token-text {
-            overflow-wrap: break-word;
-            font-family: monospace;
-        }
-        .copy-icon {
-            position: absolute;
-            top: 10px;
-            right: 10px;
-            color: #e2e8f0;
-        }
-</style>
-
-<body class="bg-gray-900 text-gray-200">
-    <div class="flex flex-col min-h-screen" x-data="networkClusters()" x-init="init()">
-        {{template "views/partials/navbar_explorer" .}}
-        <div class="animation-container">
-            <canvas id="networkCanvas"></canvas>
-            <div class="text-overlay">
-                <header class="text-center py-12">
-                    <h1 class="text-5xl font-bold text-gray-100">
-                        <i class="fa-solid fa-circle-nodes mr-2"></i> Network Clusters Explorer   
-        
-                    </h1>
-                    <p class="mt-4 text-lg">
-                        View the clusters and workers available in each network.
-                        <a href="https://localai.io/features/distribute/" target="_blank">
-                            <i class="fas fa-circle-info pr-2"></i>
-                        </a>
-                    </p>
-        
-                </header>
+    <!-- Simple navigation for login page -->
+    <nav class="bg-gray-900/80 border-b border-gray-800/60 backdrop-blur-sm">
+        <div class="container mx-auto px-4 py-4 flex justify-between items-center">
+            <div class="flex items-center">
+                <i class="fas fa-network-wired text-blue-400 text-2xl mr-3"></i>
+                <h1 class="text-xl font-bold text-white">LocalAI</h1>
             </div>
         </div>
+    </nav>
 
-        <div class="container mx-auto px-4 flex-grow">
-        <!-- Warning Box -->
-        <div class="warning-box bg-yellow-100 text-gray-800 mb-20 pt-5 pb-5 pr-5 pl-5 text-lg">
-            <i class="fa-solid fa-triangle-exclamation"></i><i class="fa-solid fa-flask"></i>
-            The explorer is a global, community-driven tool to share network tokens and view available clusters in the globe.
-            Anyone can use the tokens to offload computation and use the clusters available or share resources.
-            This is provided without any warranty. Use it at your own risk. We are not responsible for any potential harm or misuse. Sharing tokens globally allows anyone from the internet to use your instances. 
-            Although the community will address bugs, this is experimental software and may be insecure to deploy on your hardware unless you take all necessary precautions.
-        </div>
-            <div class="flow-root">
-            <!-- Toggle button for showing/hiding the form -->
-            <button class="bg-red-600 hover:bg-blue-600 float-right mb-2 flex items-center px-4 py-2 rounded" @click="toggleForm()">
-                <!-- Conditional icon display -->
-                <i :class="showForm ? 'fa-solid fa-times' : 'fa-solid fa-plus'" class="mr-2"></i>
-                <span x-text="showForm ? 'Close' : 'Add New Network'"></span>
-            </button>
-        </div>
-            <!-- Form for adding a new network -->
-            <div class="form-container" x-show="showForm" @click.outside="showForm = false">
-                <h2 class="text-3xl font-bold mb-4"><i class="fa-solid fa-plus"></i> Add New Network</h2>
-                <div class="form-control">
-                    <label for="name">Network Name</label>
-                    <input type="text" id="name" x-model="newNetwork.name" placeholder="Enter network name" />
+    <div class="container mx-auto px-4 py-8 flex-grow flex items-center justify-center">
+        <!-- Auth Card -->
+        <div class="max-w-md w-full bg-gray-800/90 border border-gray-700/50 rounded-xl overflow-hidden shadow-xl">
+            <div class="animation-container">
+                <div class="text-overlay">
+                    <i class="fas fa-circle-nodes text-5xl text-blue-400 mb-2"></i>
                 </div>
-                <div class="form-control">
-                    <label for="description">Description</label>
-                    <textarea id="description" x-model="newNetwork.description" placeholder="Enter description"></textarea>
-                </div>
-                <div class="form-control">
-                    <label for="token">Token</label>
-                    <textarea id="token" x-model="newNetwork.token" placeholder="Enter token"></textarea>
-                </div>
-                <button @click="addNetwork"><i class="fa-solid fa-plus"></i> Add Network</button>
-                <template x-if="errorMessage">
-                    <p class="error" x-text="errorMessage"></p>
-                </template>
-                <template x-if="successMessage">
-                    <p class="success" x-text="successMessage"></p>
-                </template>
             </div>
-
-            <!-- Loading Spinner -->
-            <template x-if="networks.length === 0 && !loadingComplete">
-                <div class="loading-container">
-                    <div class="spinner"></div>
-                    <p class="text-center mt-4">Loading networks...</p>
+            
+            <div class="p-8">
+                <div class="text-center mb-6">
+                    <h2 class="text-2xl font-bold text-white">
+                        <span class="bg-clip-text text-transparent bg-gradient-to-r from-blue-400 to-indigo-400">
+                            Authorization Required
+                        </span>
+                    </h2>
+                    <p class="text-gray-400 mt-2">Please enter your access token to continue</p>
                 </div>
-            </template>
-
-            <template x-if="networks.length === 0 && loadingComplete">
-                <div class="loading-container">
-                    <p class="text-center mt-4">No networks available with online workers</p>
-                </div>
-            </template>
-
-            <!-- Display Networks -->
-            <template x-for="network in networks" :key="network.name">
-                <div class="network-card">
-                    <i class="fa-solid fa-circle-nodes mr-2"></i><span class="network-title font-bold mb-4 mt-1" x-text="network.name"></span>
-                    <div class="token-box" @click="copyToken(network.token)">
-                        <p class="text-lg font-bold mb-4 mt-1">
-                            <i class="fa-solid fa-copy copy-icon"></i>
-                            <i class="fa-solid fa-key mr-2"></i>Token (click to copy): 
-                        </p>
-                        <span class="token-text" x-text="network.token"></span>
-                    </div>
-
-                    <div class="cluster">
-                        <p class="text-lg font-bold mb-4 mt-1"><i class="fa-solid fa-book mr-2"></i> Description</p>
-                        <p x-text="network.description"></p>
-                    </div>
-                    <h2 class="text-3xl font-bold mb-4 mt-4">Available Clusters in this network</h2>
-                    <template x-for="cluster in network.Clusters" :key="cluster.NetworkID + cluster.Type">
-                        <div class="cluster">
-                            <div class="cluster-title"></div>
-                            <span class="inline-block bg-orange-500 text-white py-1 px-3 rounded-full text-xs"  x-text="'Cluster Type: ' + cluster.Type">
-                            </span>
-
-                            <span class="inline-block bg-orange-500 text-white py-1 px-3 rounded-full text-xs" x-show="cluster.NetworkID" x-text="'Network ID: ' + (cluster.NetworkID || 'N/A')">
-                            </span>
-                            <span class="inline-block bg-blue-500 text-white py-1 px-3 rounded-full text-xs"  x-text="'Number of Workers: ' + cluster.Workers.length">
-                            </span>
-                            <!-- Give commands and instructions to join the network -->
-                            <span class="inline-block token-box text-white py-1 px-3 text-xs" x-show="cluster.Type == 'federated'" >
-                                <p class="text-lg font-bold mb-4 mt-1">
-                                    <i class="fa-solid fa-copy copy-icon float-right"></i>
-                                    Command to connect (click to copy): 
-                                </p>
-                                <code class="block bg-gray-700 text-yellow-300 p-4 rounded-lg break-words"  @click="copyToken($el.textContent)" >
-                                    docker run -d --restart=always -e ADDRESS=":80" -e LOCALAI_P2P_NETWORK_ID=<span class="token" x-text="cluster.NetworkID"></span> -e LOCALAI_P2P_LOGLEVEL=debug --name local-ai -e TOKEN="<span class="token" x-text="network.token"></span>" --net host -ti localai/localai:master-ffmpeg-core federated --debug
-                                </code>
-                                or via CLI:
-                                <code class="block bg-gray-700 text-yellow-300 p-4 rounded-lg break-words"  @click="copyToken($el.textContent)" >
-                                   ADDRESS=":80" LOCALAI_P2P_NETWORK_ID=<span class="token" x-text="cluster.NetworkID"></span> LOCALAI_P2P_LOGLEVEL=debug TOKEN="<span class="token" x-text="network.token"></span>" local-ai federated --debug
-                                </code>
-                            </span>
+                
+                <form id="login-form" class="space-y-6" onsubmit="login(); return false;">
+                    <div>
+                        <label for="token" class="block text-sm font-medium text-gray-300 mb-2">Access Token</label>
+                        <div class="relative">
+                            <div class="absolute inset-y-0 left-0 pl-3 flex items-center pointer-events-none">
+                                <i class="fas fa-key text-gray-500"></i>
+                            </div>
+                            <input 
+                                type="password" 
+                                id="token" 
+                                name="token" 
+                                placeholder="Enter your token" 
+                                class="bg-gray-700/50 border border-gray-600 text-white placeholder-gray-400 text-sm rounded-lg focus:ring-blue-500 focus:border-blue-500 block w-full pl-10 p-2.5"
+                                required
+                            />
                         </div>
-                    </template>
+                    </div>
+                    
+                    <div>
+                        <button 
+                            type="submit" 
+                            class="group w-full flex items-center justify-center bg-gradient-to-r from-blue-600 to-indigo-600 hover:from-blue-700 hover:to-indigo-700 text-white py-3 px-6 rounded-lg transition duration-300 ease-in-out transform hover:scale-[1.02] hover:shadow-lg font-medium"
+                        >
+                            <i class="fas fa-sign-in-alt mr-2"></i>
+                            <span>Login</span>
+                            <i class="fas fa-arrow-right opacity-0 group-hover:opacity-100 group-hover:translate-x-2 ml-2 transition-all duration-300"></i>
+                        </button>
+                    </div>
+                </form>
+                
+                <div class="mt-8 pt-6 border-t border-gray-700/50 text-center text-sm text-gray-400">
+                    <div class="flex items-center justify-center mb-2">
+                        <i class="fas fa-shield-alt mr-2 text-blue-400"></i>
+                        <span>Secure connection</span>
+                    </div>
+                    <p>Current time (UTC): <span id="current-time">{{.CurrentDate}}</span></p>
                 </div>
-            </template>
+            </div>
         </div>
-        <script>
-            function networkClusters() {
-                return {
-                    networks: [],
-                    newNetwork: {
-                        name: '',
-                        description: '',
-                        token: ''
-                    },
-                    errorMessage: '',
-                    successMessage: '',
-                    showForm: false, // Form visibility state
-                    loadingComplete: false, // To track if loading is complete
-                    toggleForm() {
-                        this.showForm = !this.showForm;
-                        console.log('Toggling form:', this.showForm);
-                    },
-                    fetchNetworks() {
-                        console.log('Fetching networks...');
-                        fetch('/networks')
-                            .then(response => response.json())
-                            .then(data => {
-                                console.log('Data fetched successfully:', data);
-                                this.networks = data;
-                                this.loadingComplete = true; // Set loading complete
-                            })
-                            .catch(error => {
-                                console.error('Error fetching networks:', error);
-                                this.loadingComplete = true; // Ensure spinner is hidden if error occurs
-                            });
-                    },
+    </div>
 
-                    addNetwork() {
-                        this.errorMessage = '';
-                        this.successMessage = '';
-                        console.log('Adding new network:', this.newNetwork);
+    {{template "views/partials/footer" .}}
+</div>
 
-                        // Validate input
-                        if (!this.newNetwork.name || !this.newNetwork.description || !this.newNetwork.token) {
-                            this.errorMessage = 'All fields are required.';
-                            return;
+<script>
+    function login() {
+        const token = document.getElementById('token').value;
+        if (!token.trim()) {
+            // Show error with fading effect
+            const form = document.getElementById('login-form');
+            const errorMsg = document.createElement('div');
+            errorMsg.className = 'p-3 mt-4 bg-red-900/50 text-red-200 rounded-lg border border-red-700/50 text-sm flex items-center';
+            errorMsg.innerHTML = '<i class="fas fa-exclamation-circle mr-2"></i> Please enter a valid token';
+            
+            // Remove any existing error message
+            const existingError = form.querySelector('.bg-red-900/50');
+            if (existingError) form.removeChild(existingError);
+            
+            // Add new error message with animation
+            form.appendChild(errorMsg);
+            setTimeout(() => {
+                errorMsg.style.opacity = '0';
+                errorMsg.style.transition = 'opacity 0.5s ease';
+                setTimeout(() => errorMsg.remove(), 500);
+            }, 3000);
+            return;
+        }
+        
+        var date = new Date();
+        date.setTime(date.getTime() + (24*60*60*1000));
+        document.cookie = `token=${token}; expires=${date.toGMTString()}; path=/`;
+
+        // Show loading state
+        const button = document.querySelector('button[type="submit"]');
+        const originalContent = button.innerHTML;
+        button.disabled = true;
+        button.innerHTML = '<i class="fas fa-spinner fa-spin mr-2"></i> Authenticating...';
+        button.classList.add('bg-gray-600');
+        
+        // Reload after short delay to show loading state
+        setTimeout(() => {
+            window.location.reload();
+        }, 800);
+    }
+    
+    // Update current time
+    function updateCurrentTime() {
+        const timeElement = document.getElementById('current-time');
+        if (timeElement) {
+            const now = new Date();
+            const year = now.getUTCFullYear();
+            const month = String(now.getUTCMonth() + 1).padStart(2, '0');
+            const day = String(now.getUTCDate()).padStart(2, '0');
+            const hours = String(now.getUTCHours()).padStart(2, '0');
+            const minutes = String(now.getUTCMinutes()).padStart(2, '0');
+            const seconds = String(now.getUTCSeconds()).padStart(2, '0');
+            timeElement.textContent = `${year}-${month}-${day} ${hours}:${minutes}:${seconds}`;
+        }
+    }
+    
+    // Initialize current time and update it every second
+    updateCurrentTime();
+    setInterval(updateCurrentTime, 1000);
+    
+    // Add subtle particle animation to the background
+    document.addEventListener('DOMContentLoaded', function() {
+        const animContainer = document.querySelector('.animation-container');
+        if (animContainer) {
+            const canvas = document.createElement('canvas');
+            animContainer.appendChild(canvas);
+            
+            const ctx = canvas.getContext('2d');
+            canvas.width = animContainer.offsetWidth;
+            canvas.height = animContainer.offsetHeight;
+            
+            // Create particles
+            const particles = [];
+            const particleCount = 30;
+            
+            for (let i = 0; i < particleCount; i++) {
+                particles.push({
+                    x: Math.random() * canvas.width,
+                    y: Math.random() * canvas.height,
+                    radius: Math.random() * 3 + 1,
+                    color: `rgba(${Math.random() * 50 + 50}, ${Math.random() * 100 + 100}, ${Math.random() * 155 + 100}, ${Math.random() * 0.4 + 0.1})`,
+                    speedX: Math.random() * 0.5 - 0.25,
+                    speedY: Math.random() * 0.5 - 0.25
+                });
+            }
+            
+            // Animation loop
+            function animate() {
+                requestAnimationFrame(animate);
+                ctx.clearRect(0, 0, canvas.width, canvas.height);
+                
+                particles.forEach(particle => {
+                    particle.x += particle.speedX;
+                    particle.y += particle.speedY;
+                    
+                    // Bounce off edges
+                    if (particle.x < 0 || particle.x > canvas.width) {
+                        particle.speedX = -particle.speedX;
+                    }
+                    
+                    if (particle.y < 0 || particle.y > canvas.height) {
+                        particle.speedY = -particle.speedY;
+                    }
+                    
+                    // Draw particle
+                    ctx.beginPath();
+                    ctx.arc(particle.x, particle.y, particle.radius, 0, Math.PI * 2);
+                    ctx.fillStyle = particle.color;
+                    ctx.fill();
+                });
+                
+                // Connect nearby particles with lines
+                for (let i = 0; i < particles.length; i++) {
+                    for (let j = i + 1; j < particles.length; j++) {
+                        const dx = particles[i].x - particles[j].x;
+                        const dy = particles[i].y - particles[j].y;
+                        const distance = Math.sqrt(dx * dx + dy * dy);
+                        
+                        if (distance < 100) {
+                            ctx.beginPath();
+                            ctx.moveTo(particles[i].x, particles[i].y);
+                            ctx.lineTo(particles[j].x, particles[j].y);
+                            ctx.strokeStyle = `rgba(100, 150, 255, ${0.1 * (1 - distance / 100)})`;
+                            ctx.lineWidth = 1;
+                            ctx.stroke();
                         }
-
-                        fetch('/network/add', {
-                            method: 'POST',
-                            headers: {
-                                'Content-Type': 'application/json'
-                            },
-                            body: JSON.stringify(this.newNetwork)
-                        })
-                            .then(response => {
-                                if (!response.ok) {
-                                    return response.json().then(err => { throw err; });
-                                }
-                                return response.json();
-                            })
-                            .then(data => {
-                                console.log('Network added successfully:', data);
-                                this.successMessage = 'Network added successfully!';
-                                this.fetchNetworks(); // Refresh the networks list
-                                this.newNetwork = { name: '', description: '', token: '' }; // Clear form
-                            })
-                            .catch(error => {
-                                console.error('Error adding network:', error);
-                                this.errorMessage = 'Failed to add network. Please try again.'
-                                if (error.error) {
-                                    this.errorMessage += " Error : " + error.error;
-                                }
-                            });
-                    },
-                    copyToken(token) {
-                        navigator.clipboard.writeText(token)
-                        .then(() => {
-                            console.log('Text copied to clipboard:', token);
-                            alert('Text copied to clipboard!');
-                        })
-                        .catch(err => {
-                            console.error('Failed to copy token:', err);
-                        });
-                    },
-                    init() {
-                        console.log('Initializing Alpine component...');
-                        this.fetchNetworks();
-                        setInterval(() => {
-                            this.fetchNetworks();
-                        }, 5000); // Refresh every 5 seconds
                     }
                 }
             }
-        </script>
-        <script src="static/p2panimation.js"></script>
-
-        {{template "views/partials/footer" .}}
-    </div>
+            
+            // Start animation
+            animate();
+            
+            // Resize handling
+            window.addEventListener('resize', () => {
+                canvas.width = animContainer.offsetWidth;
+                canvas.height = animContainer.offsetHeight;
+            });
+        }
+    });
+</script>
 
 </body>
-
-</html>
+</html>

core/http/views/index.html

@@ -2,113 +2,185 @@
 <html lang="en">
 {{template "views/partials/head" .}}
 
-<body class="bg-gray-900 text-gray-200">
+<body class="bg-gradient-to-br from-gray-900 to-gray-950 text-gray-200">
 <div class="flex flex-col min-h-screen">
 
     {{template "views/partials/navbar" .}}
 
-    <div class="container mx-auto px-4 flex-grow">
-        <div class="header text-center py-12">
-            <h1 class="text-5xl font-bold text-gray-100">Welcome to <i>your</i> LocalAI instance!</h1>
-            <p class="mt-4 text-lg">The FOSS alternative to OpenAI, Claude, ...</p>
-            <a href="https://localai.io" target="_blank" class="mt-4 inline-block bg-blue-500 text-white py-2 px-4 rounded-lg shadow transition duration-300 ease-in-out hover:bg-blue-700 hover:shadow-lg">
-                <i class="fas fa-book-reader pr-2"></i>Documentation
-            </a>
+    <div class="container mx-auto px-4 py-8 flex-grow">
+        <!-- Hero Section -->
+        <div class="bg-gradient-to-r from-blue-900/30 to-indigo-900/30 rounded-2xl shadow-xl p-8 mb-10">
+            <div class="max-w-4xl mx-auto text-center">
+                <h1 class="text-4xl md:text-5xl font-bold text-white mb-4">
+                    <span class="bg-clip-text text-transparent bg-gradient-to-r from-blue-400 to-indigo-400">
+                        Welcome to <i>your</i> LocalAI instance!
+                    </span>
+                </h1>
+                <p class="text-xl text-gray-300 mb-6">The FOSS alternative to OpenAI, Claude, and more</p>
+                <div class="flex flex-wrap justify-center gap-4">
+                    <a href="https://localai.io" target="_blank" 
+                       class="group flex items-center bg-blue-600 hover:bg-blue-700 text-white py-2 px-6 rounded-lg transition duration-300 ease-in-out transform hover:scale-105 hover:shadow-lg">
+                        <i class="fas fa-book-reader mr-2"></i>
+                        <span>Documentation</span>
+                        <i class="fas fa-arrow-right opacity-0 group-hover:opacity-100 group-hover:translate-x-2 ml-2 transition-all duration-300"></i>
+                    </a>
+                    <a href="browse" 
+                       class="group flex items-center bg-indigo-600 hover:bg-indigo-700 text-white py-2 px-6 rounded-lg transition duration-300 ease-in-out transform hover:scale-105 hover:shadow-lg">
+                        <i class="fas fa-images mr-2"></i>
+                        <span>Gallery</span>
+                        <i class="fas fa-arrow-right opacity-0 group-hover:opacity-100 group-hover:translate-x-2 ml-2 transition-all duration-300"></i>
+                    </a>
+                </div>
+            </div>
         </div>
 
-        <div class="models mt-4">
+        <!-- Models Section -->
+        <div class="models mt-8">
             {{template "views/partials/inprogress" .}}
+            
             {{ if eq (len .ModelsConfig) 0 }}
-            <h2 class="text-center text-3xl font-semibold text-gray-100"> <i class="text-yellow-200 ml-2 fa-solid fa-triangle-exclamation animate-pulse"></i> Ouch! seems you don't have any models installed from the LocalAI gallery!</h2>
-            <p class="text-center mt-4 text-xl">..install something from the <a class="text-gray-400 hover:text-white ml-1 px-3 py-2 rounded" href="browse">🖼️ Gallery</a> or check the <a href="https://localai.io/basics/getting_started/" class="text-gray-400 hover:text-white ml-1 px-3 py-2 rounded"> <i class="fa-solid fa-book"></i> Getting started documentation </a></p>
+            <div class="bg-gray-800/50 border border-gray-700/50 rounded-xl p-8 shadow-md backdrop-blur-sm">
+                <div class="text-center max-w-3xl mx-auto">
+                    <div class="inline-flex items-center justify-center w-16 h-16 rounded-full bg-yellow-500/20 mb-4">
+                        <i class="text-yellow-400 text-2xl fa-solid fa-triangle-exclamation"></i>
+                    </div>
+                    <h2 class="text-2xl md:text-3xl font-semibold text-gray-100 mb-4">No models installed from the LocalAI gallery</h2>
+                    <p class="text-lg text-gray-300 mb-6">Install models from the <a class="text-blue-400 hover:text-blue-300 underline underline-offset-2" href="browse">🖼️ Gallery</a> or check the <a href="https://localai.io/basics/getting_started/" class="text-blue-400 hover:text-blue-300 underline underline-offset-2"> <i class="fa-solid fa-book"></i> Getting started documentation</a></p>
 
-            {{ if ne (len .Models) 0 }}
-            <hr class="my-4">
-            <h3 class="text-center text-xl font-semibold text-gray-100"> 
-                However, It seems you have installed some models installed without a configuration file:
-            </h3>
-            {{ range .Models }}
-            <div class="bg-gray-800 border-b border-gray-700 p-4 mt-4">
-                <h4 class="text-md font-bold text-gray-200">{{.}}</h4>
-            </div>              
-            {{end}}
-            {{end}}
+                    {{ if ne (len .Models) 0 }}
+                    <div class="mt-8 pt-8 border-t border-gray-700/50">
+                        <h3 class="text-xl font-semibold text-gray-100 mb-4">Models installed without a configuration file:</h3>
+                        <div class="grid grid-cols-1 md:grid-cols-2 lg:grid-cols-3 gap-4">
+                            {{ range .Models }}
+                            <div class="bg-gray-800/80 border border-gray-700 rounded-lg p-4 flex items-center">
+                                <i class="fas fa-brain text-lg text-gray-400 mr-3"></i>
+                                <p class="font-medium text-gray-200">{{.Name}}</p>
+                            </div>              
+                            {{end}}
+                        </div>
+                    </div>
+                    {{end}}
+                </div>
+            </div>
             {{ else }}
             {{ $modelsN := len .ModelsConfig}}
             {{ $modelsN = add $modelsN (len .Models)}}
-            <h2 class="text-center text-3xl font-semibold text-gray-100">{{$modelsN}} Installed model(s)</h2>
-            <table class="table-auto mt-4 w-full text-left text-gray-200">
-                <thead class="text-xs text-gray-400 uppercase bg-gray-700">
-                    <tr>
-                        <th class="px-4 py-2"></th>
-                        <th class="px-4 py-2">Model Name</th>
-                        <th class="px-4 py-2">Backend</th>
-                        <th class="px-4 py-2 float-right">Actions</th>
-                    </tr>
-                </thead>
-                <tbody>
+            <div class="mb-6 flex flex-col md:flex-row md:items-center md:justify-between">
+                <h2 class="text-2xl md:text-3xl font-bold text-white mb-4 md:mb-0">
+                    <span class="text-blue-400">{{$modelsN}}</span> Installed Model<span class="{{if gt $modelsN 1}}s{{end}}">
+                </h2>
+                <!--
+                <div class="flex gap-4">
+                    <button class="text-sm bg-gray-800 hover:bg-gray-700 text-gray-300 py-2 px-4 rounded-lg transition flex items-center gap-2">
+                        <i class="fas fa-filter"></i> Filter
+                    </button>
+                    <button class="text-sm bg-gray-800 hover:bg-gray-700 text-gray-300 py-2 px-4 rounded-lg transition flex items-center gap-2">
+                        <i class="fas fa-sort"></i> Sort
+                    </button>
+                </div>
+                -->
+            </div>
+            
+            <div class="grid grid-cols-1 md:grid-cols-2 lg:grid-cols-3 gap-6">
                 {{$galleryConfig:=.GalleryConfig}}
                 {{$noicon:="https://upload.wikimedia.org/wikipedia/commons/6/65/No-Image-Placeholder.svg"}}
+                
                 {{ range .ModelsConfig }}
+                {{ $backendCfg := . }}
                 {{ $cfg:= index $galleryConfig .Name}}
-                <tr class="bg-gray-800 border-b border-gray-700">
-                    <td class="px-4 py-3">
-                        {{ with $cfg }}
-                        <img {{ if $cfg.Icon }}
-                            src="{{$cfg.Icon}}"
-                            {{ else }}
-                            src="{{$noicon}}"
-                            {{ end }}
-                            class="rounded-t-lg max-h-24 max-w-24 object-cover mt-3"
+                <div class="bg-gray-800/90 border border-gray-700/50 rounded-xl overflow-hidden transition-all duration-300 hover:shadow-lg hover:shadow-blue-900/20 hover:-translate-y-1 hover:border-blue-700/50">
+                    <div class="flex p-5">
+                        <div class="w-20 h-20 rounded-lg overflow-hidden flex-shrink-0 bg-gray-700/50 flex items-center justify-center">
+                            <img {{ if and $cfg $cfg.Icon }}
+                                src="{{$cfg.Icon}}"
+                                {{ else }}
+                                src="{{$noicon}}"
+                                {{ end }}
+                                class="w-full h-full object-contain"
+                                alt="{{.Name}} icon"
                             >
-                        {{ else}}
-                            <img src="{{$noicon}}" class="rounded-t-lg max-h-24 max-w-24 object-cover mt-3">
-                        {{ end }}
-                    </td>
-                    <td class="px-4 py-3 font-bold">
-                        <p class="font-bold text-white flex items-center"><i class="fas fa-brain pr-2"></i><a href="browse?term={{.Name}}">{{.Name}}</a></p>
-                    </td>
-                    <td class="px-4 py-3 font-bold">
-                        {{ if .Backend }}
-                        <!-- Badge for Backend -->
-                        <span class="inline-block bg-blue-500 text-white py-1 px-3 rounded-full text-xs">
-                            {{.Backend}}
-                        </span>
-                        {{ else }}
-                        <span class="inline-block bg-yellow-500 text-white py-1 px-3 rounded-full text-xs">
-                            auto
-                        </span>
-                        {{ end }}
-                    </td>
-
-                    <td class="px-4 py-3">
-                        <button
-                            class="float-right inline-block rounded bg-red-800 px-6 pb-2.5 mb-3 pt-2.5 text-xs font-medium uppercase leading-normal text-white shadow-primary-3 transition duration-150 ease-in-out hover:bg-red-accent-300 hover:shadow-red-2 focus:bg-red-accent-300 focus:shadow-primary-2 focus:outline-none focus:ring-0 active:bg-red-600 active:shadow-primary-2 dark:shadow-black/30 dark:hover:shadow-dark-strong dark:focus:shadow-dark-strong dark:active:shadow-dark-strong"
-                            data-twe-ripple-color="light" data-twe-ripple-init="" hx-confirm="Are you sure you wish to delete the model?" hx-post="browse/delete/model/{{.Name}}" hx-swap="outerHTML"><i class="fa-solid fa-cancel pr-2"></i>Delete</button>
-                    </td>
+                        </div>
+                        <div class="ml-4 flex-1 min-w-0">
+                            <div class="flex items-center">
+                                <h3 class="font-bold text-lg text-white truncate">{{.Name}}</h3>
+                                <a href="browse?term={{.Name}}" class="ml-2 text-gray-400 hover:text-blue-400 transition" title="Search for similar models">
+                                    <i class="fas fa-search text-xs"></i>
+                                </a>
+                            </div>
+                            
+                            <div class="mt-2 flex flex-wrap gap-2">
+                                {{ if .Backend }}
+                                <span class="inline-flex items-center px-2.5 py-0.5 rounded-md text-xs font-medium bg-blue-900/50 text-blue-300 border border-blue-700/50">
+                                    {{.Backend}}
+                                </span>
+                                {{ else }}
+                                <span class="inline-flex items-center px-2.5 py-0.5 rounded-md text-xs font-medium bg-yellow-900/50 text-yellow-300 border border-yellow-700/50">
+                                    auto
+                                </span>
+                                {{ end }}
+                            </div>
+                        </div>
+                    </div>
+                    
+                    <div class="px-5 pb-5 pt-2">
+                        <div class="flex flex-wrap gap-2">
+                            {{ range .KnownUsecaseStrings }}
+                            {{ if eq . "FLAG_CHAT" }}
+                                <a href="chat/{{$backendCfg.Name}}" class="inline-flex items-center rounded-full px-4 py-2 text-sm font-medium bg-blue-900/60 text-blue-200 border border-blue-700/50 hover:bg-blue-800 transition duration-200 ease-in-out">
+                                    <i class="fas fa-comment-alt text-xs mr-1.5"></i>Chat
+                                </a>
+                            {{ end }}
+                            {{ if eq . "FLAG_IMAGE" }}
+                                <a href="text2image/{{$backendCfg.Name}}" class="inline-flex items-center text-sm bg-green-600/80 hover:bg-green-700 text-white py-1.5 px-3 rounded-lg shadow transition duration-300 ease-in-out">
+                                    <i class="fas fa-image text-xs mr-1.5"></i>Image
+                                </a>
+                            {{ end }}
+                            {{ if eq . "FLAG_TTS" }}
+                                <a href="tts/{{$backendCfg.Name}}" class="inline-flex items-center text-sm bg-purple-600/80 hover:bg-purple-700 text-white py-1.5 px-3 rounded-lg shadow transition duration-300 ease-in-out">
+                                    <i class="fas fa-microphone text-xs mr-1.5"></i>TTS
+                                </a>
+                            {{ end }}
+                            {{ end }}
+                        </div>
+                        
+                        <div class="mt-4 flex justify-end">
+                            <button
+                                class="inline-flex items-center text-xs font-medium text-red-400 hover:text-red-300 hover:bg-red-900/20 rounded-md px-2 py-1 transition-colors duration-200"
+                                data-twe-ripple-init=""
+                                hx-confirm="Are you sure you wish to delete this model?"
+                                hx-post="browse/delete/model/{{.Name}}"
+                                hx-swap="outerHTML">
+                                <i class="fas fa-trash-alt mr-1.5"></i>Delete
+                            </button>
+                        </div>
+                    </div>
+                </div>
                 {{ end }}
+                
                 {{ range .Models }}
-                <tr class="bg-gray-800 border-b border-gray-700">
-                    <td class="px-4 py-3">
-                            <img src="{{$noicon}}" class="rounded-t-lg max-h-24 max-w-24 object-cover mt-3">
-                    </td>
-                    <td class="px-4 py-3 font-bold">
-                        <p class="font-bold text-white flex items-center"><i class="fas fa-brain pr-2"></i>{{.}}</p>
-                    </td>
-                    <td class="px-4 py-3 font-bold">
-                        <span class="inline-block bg-yellow-500 text-white py-1 px-3 rounded-full text-xs">
-                            auto
-                        </span>
-                    </td>
-
-                    <td class="px-4 py-3">
-                        <span class="float-right inline-block bg-red-800 text-white py-1 px-3 rounded-full text-xs">
-                            No Configuration
-                        </span>
-                    </td>           
+                <div class="bg-gray-800/90 border border-gray-700/50 rounded-xl overflow-hidden transition-all duration-300 hover:shadow-lg hover:shadow-blue-900/20 hover:-translate-y-1 hover:border-blue-700/50">
+                    <div class="flex p-5">
+                        <div class="w-20 h-20 rounded-lg overflow-hidden flex-shrink-0 bg-gray-700/50 flex items-center justify-center">
+                            <img src="{{$noicon}}" class="w-full h-full object-contain" alt="Model icon">
+                        </div>
+                        <div class="ml-4 flex-1 min-w-0">
+                            <div class="flex items-center">
+                                <h3 class="font-bold text-lg text-white truncate"><i class="fas fa-brain mr-2 text-gray-400"></i>{{.}}</h3>
+                            </div>
+                            
+                            <div class="mt-2 flex flex-wrap gap-2">
+                                <span class="inline-flex items-center px-2.5 py-0.5 rounded-md text-xs font-medium bg-yellow-900/50 text-yellow-300 border border-yellow-700/50">
+                                    auto
+                                </span>
+                                <span class="inline-flex items-center px-2.5 py-0.5 rounded-md text-xs font-medium bg-red-900/50 text-red-300 border border-red-700/50">
+                                    No Configuration
+                                </span>
+                            </div>
+                        </div>
+                    </div>
+                </div>
                 {{end}}
-                </tbody>
-            </table>
+            </div>
             {{ end }}
         </div>
     </div>
@@ -117,4 +189,4 @@
 </div>
 
 </body>
-</html>
+</html>

core/http/views/login.html

@@ -1,25 +1,216 @@
 <!DOCTYPE html>
 <html lang="en">
-<head>
-    <meta charset="UTF-8">
-    <meta name="viewport" content="width=device-width, initial-scale=1.0">
-    <title>Open Authenticated Website</title>
-    <base href="{{.BaseURL}}" />
-    <link rel="icon" type="image/x-icon" href="favicon.ico" />
-</head>
-<body>
-    <h1>Authorization is required</h1>
-    <input type="text" id="token" placeholder="Token" />
-    <button onclick="login()">Login</button>
-    <script>
-        function login() {
-            const token = document.getElementById('token').value;
-            var date = new Date();
-            date.setTime(date.getTime() + (24*60*60*1000));
-            document.cookie = `token=${token}; expires=${date.toGMTString()}`;
+{{template "views/partials/head" .}}
 
-            window.location.reload();
+<body class="bg-gradient-to-br from-gray-900 to-gray-950 text-gray-200">
+<div class="flex flex-col min-h-screen">
+
+    {{template "views/partials/navbar" .}}
+
+    <div class="container mx-auto px-4 py-8 flex-grow flex items-center justify-center">
+        <!-- Auth Card -->
+        <div class="max-w-md w-full bg-gray-800/90 border border-gray-700/50 rounded-xl overflow-hidden shadow-xl">
+            <div class="animation-container">
+                <div class="text-overlay">
+                <!--    <i class="fas fa-circle-nodes text-5xl text-blue-400 mb-2"></i> -->
+                </div>
+            </div>
+            
+            <div class="p-8">
+                <div class="text-center mb-6">
+                    <h2 class="text-2xl font-bold text-white">
+                        <span class="bg-clip-text text-transparent bg-gradient-to-r from-blue-400 to-indigo-400">
+                            Authorization Required
+                        </span>
+                    </h2>
+                    <p class="text-gray-400 mt-2">Please enter your access token to continue</p>
+                </div>
+                
+                <form id="login-form" class="space-y-6" onsubmit="login(); return false;">
+                    <div>
+                        <label for="token" class="block text-sm font-medium text-gray-300 mb-2">Access Token</label>
+                        <div class="relative">
+                            <div class="absolute inset-y-0 left-0 pl-3 flex items-center pointer-events-none">
+                                <i class="fas fa-key text-gray-500"></i>
+                            </div>
+                            <input 
+                                type="password" 
+                                id="token" 
+                                name="token" 
+                                placeholder="Enter your token" 
+                                class="bg-gray-700/50 border border-gray-600 text-white placeholder-gray-400 text-sm rounded-lg focus:ring-blue-500 focus:border-blue-500 block w-full pl-10 p-2.5"
+                                required
+                            />
+                        </div>
+                    </div>
+                    
+                    <div>
+                        <button 
+                            type="submit" 
+                            class="group w-full flex items-center justify-center bg-gradient-to-r from-blue-600 to-indigo-600 hover:from-blue-700 hover:to-indigo-700 text-white py-3 px-6 rounded-lg transition duration-300 ease-in-out transform hover:scale-[1.02] hover:shadow-lg font-medium"
+                        >
+                            <i class="fas fa-sign-in-alt mr-2"></i>
+                            <span>Login</span>
+                            <i class="fas fa-arrow-right opacity-0 group-hover:opacity-100 group-hover:translate-x-2 ml-2 transition-all duration-300"></i>
+                        </button>
+                    </div>
+                </form>
+                
+                <div class="mt-8 pt-6 border-t border-gray-700/50 text-center text-sm text-gray-400">
+                    <div class="flex items-center justify-center mb-2">
+                        <i class="fas fa-shield-alt mr-2 text-blue-400"></i>
+                        <span>Instance is token protected</span>
+                    </div>
+                    <p>Current time (UTC): <span id="current-time">{{.CurrentDate}}</span></p>
+                </div>
+            </div>
+        </div>
+    </div>
+
+    {{template "views/partials/footer" .}}
+</div>
+
+<script>
+    function login() {
+        const token = document.getElementById('token').value;
+        if (!token.trim()) {
+            // Show error with fading effect
+            const form = document.getElementById('login-form');
+            const errorMsg = document.createElement('div');
+            errorMsg.className = 'p-3 mt-4 bg-red-900/50 text-red-200 rounded-lg border border-red-700/50 text-sm flex items-center';
+            errorMsg.innerHTML = '<i class="fas fa-exclamation-circle mr-2"></i> Please enter a valid token';
+            
+            // Remove any existing error message
+            const existingError = form.querySelector('.bg-red-900/50');
+            if (existingError) form.removeChild(existingError);
+            
+            // Add new error message with animation
+            form.appendChild(errorMsg);
+            setTimeout(() => {
+                errorMsg.style.opacity = '0';
+                errorMsg.style.transition = 'opacity 0.5s ease';
+                setTimeout(() => errorMsg.remove(), 500);
+            }, 3000);
+            return;
         }
-    </script>
+        
+        var date = new Date();
+        date.setTime(date.getTime() + (24*60*60*1000));
+        document.cookie = `token=${token}; expires=${date.toGMTString()}; path=/`;
+
+        // Show loading state
+        const button = document.querySelector('button[type="submit"]');
+        const originalContent = button.innerHTML;
+        button.disabled = true;
+        button.innerHTML = '<i class="fas fa-spinner fa-spin mr-2"></i> Authenticating...';
+        button.classList.add('bg-gray-600');
+        
+        // Reload after short delay to show loading state
+        setTimeout(() => {
+            window.location.reload();
+        }, 800);
+    }
+    
+    // Update current time
+    function updateCurrentTime() {
+        const timeElement = document.getElementById('current-time');
+        if (timeElement) {
+            const now = new Date();
+            const year = now.getUTCFullYear();
+            const month = String(now.getUTCMonth() + 1).padStart(2, '0');
+            const day = String(now.getUTCDate()).padStart(2, '0');
+            const hours = String(now.getUTCHours()).padStart(2, '0');
+            const minutes = String(now.getUTCMinutes()).padStart(2, '0');
+            const seconds = String(now.getUTCSeconds()).padStart(2, '0');
+            timeElement.textContent = `${year}-${month}-${day} ${hours}:${minutes}:${seconds}`;
+        }
+    }
+    
+    // Initialize current time and update it every second
+    updateCurrentTime();
+    setInterval(updateCurrentTime, 1000);
+    
+    // Add subtle particle animation to the background
+    document.addEventListener('DOMContentLoaded', function() {
+        const animContainer = document.querySelector('.animation-container');
+        if (animContainer) {
+            const canvas = document.createElement('canvas');
+            animContainer.appendChild(canvas);
+            
+            const ctx = canvas.getContext('2d');
+            canvas.width = animContainer.offsetWidth;
+            canvas.height = animContainer.offsetHeight;
+            
+            // Create particles
+            const particles = [];
+            const particleCount = 30;
+            
+            for (let i = 0; i < particleCount; i++) {
+                particles.push({
+                    x: Math.random() * canvas.width,
+                    y: Math.random() * canvas.height,
+                    radius: Math.random() * 3 + 1,
+                    color: `rgba(${Math.random() * 50 + 50}, ${Math.random() * 100 + 100}, ${Math.random() * 155 + 100}, ${Math.random() * 0.4 + 0.1})`,
+                    speedX: Math.random() * 0.5 - 0.25,
+                    speedY: Math.random() * 0.5 - 0.25
+                });
+            }
+            
+            // Animation loop
+            function animate() {
+                requestAnimationFrame(animate);
+                ctx.clearRect(0, 0, canvas.width, canvas.height);
+                
+                particles.forEach(particle => {
+                    particle.x += particle.speedX;
+                    particle.y += particle.speedY;
+                    
+                    // Bounce off edges
+                    if (particle.x < 0 || particle.x > canvas.width) {
+                        particle.speedX = -particle.speedX;
+                    }
+                    
+                    if (particle.y < 0 || particle.y > canvas.height) {
+                        particle.speedY = -particle.speedY;
+                    }
+                    
+                    // Draw particle
+                    ctx.beginPath();
+                    ctx.arc(particle.x, particle.y, particle.radius, 0, Math.PI * 2);
+                    ctx.fillStyle = particle.color;
+                    ctx.fill();
+                });
+                
+                // Connect nearby particles with lines
+                for (let i = 0; i < particles.length; i++) {
+                    for (let j = i + 1; j < particles.length; j++) {
+                        const dx = particles[i].x - particles[j].x;
+                        const dy = particles[i].y - particles[j].y;
+                        const distance = Math.sqrt(dx * dx + dy * dy);
+                        
+                        if (distance < 100) {
+                            ctx.beginPath();
+                            ctx.moveTo(particles[i].x, particles[i].y);
+                            ctx.lineTo(particles[j].x, particles[j].y);
+                            ctx.strokeStyle = `rgba(100, 150, 255, ${0.1 * (1 - distance / 100)})`;
+                            ctx.lineWidth = 1;
+                            ctx.stroke();
+                        }
+                    }
+                }
+            }
+            
+            // Start animation
+            animate();
+            
+            // Resize handling
+            window.addEventListener('resize', () => {
+                canvas.width = animContainer.offsetWidth;
+                canvas.height = animContainer.offsetHeight;
+            });
+        }
+    });
+</script>
+
 </body>
-</html>
+</html>

core/http/views/models.html

@@ -2,84 +2,185 @@
 <html lang="en">
 {{template "views/partials/head" .}}
 
-<body class="bg-gray-900 text-gray-200">
+<body class="bg-gradient-to-br from-gray-900 to-gray-950 text-gray-200">
 <div class="flex flex-col min-h-screen">
    
     {{template "views/partials/navbar" .}}
-    <div class="container mx-auto px-4 flex-grow">
+    {{ $numModelsPerPage := 21 }}
+    <div class="container mx-auto px-4 py-8 flex-grow">
 
-        <div class="models mt-12">
-            <h2 class="text-center text-3xl font-semibold text-gray-100">
-                🖼️ Available {{.AvailableModels}} models from <i>{{ len .Repositories }}</i> repositories     <a href="https://localai.io/models/" target="_blank" >
-                    <i class="fas fa-circle-info pr-2"></i>
-                </a></h2> 
-
-            <div class="text-center font-semibold text-gray-100">
-                <h2>Filter by type:</h2>
-                <button  hx-post="browse/search/models"
-                    class="text-white-500 inline-block bg-blue-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2 hover:bg-gray-300 hover:shadow-gray-2"
-                    hx-target="#search-results" 
-                    hx-vals='{"search": "tts"}'
-                hx-indicator=".htmx-indicator" >TTS</button> 
-                <button  hx-post="browse/search/models" 
-                    class="text-white-500 inline-block bg-blue-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2 hover:bg-gray-300 hover:shadow-gray-2"
-                    hx-target="#search-results" 
-                    hx-vals='{"search": "stablediffusion"}'
-                hx-indicator=".htmx-indicator" >Image generation</button> 
-                <button  hx-post="browse/search/models" \
-                    class="text-white-500 inline-block bg-blue-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2 hover:bg-gray-300 hover:shadow-gray-2"
-                    hx-target="#search-results" 
-                    hx-vals='{"search": "llm"}'
-                hx-indicator=".htmx-indicator" >Text generation</button> 
-                <button  hx-post="browse/search/models" 
-                    class="text-white-500 inline-block bg-blue-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2 hover:bg-gray-300 hover:shadow-gray-2"
-                    hx-target="#search-results" 
-                    hx-vals='{"search": "multimodal"}'
-                hx-indicator=".htmx-indicator" >Multimodal</button> 
-                <button  hx-post="browse/search/models" 
-                    class="text-white-500 inline-block bg-blue-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2 hover:bg-gray-300 hover:shadow-gray-2"
-                    hx-target="#search-results" 
-                    hx-vals='{"search": "embedding"}'
-                hx-indicator=".htmx-indicator" >Embeddings</button>
-                <button  hx-post="browse/search/models"
-                    class="text-white-500 inline-block bg-blue-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2 hover:bg-gray-300 hover:shadow-gray-2"
-                    hx-target="#search-results" 
-                    hx-vals='{"search": "rerank"}'
-                hx-indicator=".htmx-indicator" >Rerankers</button> 
-                <button  
-                    hx-post="browse/search/models"
-                    class="text-white-500 inline-block bg-blue-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2 hover:bg-gray-300 hover:shadow-gray-2"
-                    hx-target="#search-results" 
-                    hx-vals='{"search": "whisper"}'
-                hx-indicator=".htmx-indicator" >Audio transcription</button> 
+        <!-- Hero Header -->
+        <div class="bg-gradient-to-r from-indigo-900/30 to-purple-900/30 rounded-2xl shadow-xl p-6 mb-8">
+            <div class="max-w-4xl mx-auto text-center">
+                <h1 class="text-3xl md:text-4xl font-bold text-white mb-3">
+                    <span class="bg-clip-text text-transparent bg-gradient-to-r from-indigo-400 to-purple-400">
+                        Model Gallery
+                    </span>
+                </h1>
+                <p class="text-lg text-gray-300 mb-2">
+                    <span class="font-semibold text-indigo-300">{{.AvailableModels}}</span> models from 
+                    <span class="font-semibold text-purple-300">{{ len .Repositories }}</span> repositories
+                    <a href="https://localai.io/models/" target="_blank" class="ml-2 text-blue-400 hover:text-blue-300 transition">
+                        <i class="fas fa-circle-info"></i>
+                    </a>
+                </p>
             </div>
-
-            <div class="text-center text-xs font-semibold text-gray-100">
-                Filter by tags:
-                {{ range .AllTags }}
-                    <button  hx-post="browse/search/models" class="text-blue-500" hx-target="#search-results" 
-                    hx-vals='{"search": "{{.}}"}'
-                    hx-indicator=".htmx-indicator" >{{.}}</button> 
-                {{ end }}
-            </div>
-
-            
-            <span class="htmx-indicator loader"></span>
-            {{template "views/partials/inprogress" .}}
-
-            <input class="form-control appearance-none block w-full mt-5 px-3 py-2 text-base font-normal text-gray-300 pb-2 mb-5 bg-gray-800 bg-clip-padding border border-solid border-gray-600 rounded transition ease-in-out m-0 focus:text-gray-300 focus:bg-gray-900 focus:border-blue-500 focus:outline-none" type="search" 
-                name="search" placeholder="Begin Typing To Search models..." 
-                hx-post="browse/search/models" 
-                hx-trigger="input changed delay:500ms, search" 
-                hx-target="#search-results" 
-                hx-indicator=".htmx-indicator">
-
-            <div id="search-results">{{.Models}}</div>
         </div>
-    </div>
+        
+        {{template "views/partials/inprogress" .}}
 
+        <!-- Search and Filter Section -->
+        <div class="bg-gray-800/70 rounded-xl p-6 mb-8 shadow-lg border border-gray-700/50">
+            <!-- Search Input -->
+            <div class="relative mb-6">
+                <div class="absolute inset-y-0 start-0 flex items-center ps-3 pointer-events-none">
+                    <i class="fas fa-search text-gray-400"></i>
+                </div>
+                <input class="form-control block w-full pl-10 px-4 py-3 text-base font-normal text-gray-300 bg-gray-900/80 bg-clip-padding border border-gray-700/70 rounded-lg transition ease-in-out focus:text-gray-200 focus:bg-gray-900 focus:border-blue-500 focus:ring-1 focus:ring-blue-500/50 focus:outline-none" 
+                    type="search" 
+                    name="search" 
+                    placeholder="Search models by name, tag, or description..." 
+                    hx-post="browse/search/models" 
+                    hx-trigger="input changed delay:500ms, search" 
+                    hx-target="#search-results"
+                    oninput="hidePagination()"
+                    onchange="hidePagination()"
+                    onsearch="hidePagination()"
+                    hx-indicator=".htmx-indicator">
+                <span class="htmx-indicator absolute right-3 top-3">
+                    <svg class="animate-spin h-5 w-5 text-blue-500" xmlns="http://www.w3.org/2000/svg" fill="none" viewBox="0 0 24 24">
+                        <circle class="opacity-25" cx="12" cy="12" r="10" stroke="currentColor" stroke-width="4"></circle>
+                        <path class="opacity-75" fill="currentColor" d="M4 12a8 8 0 018-8V0C5.373 0 0 5.373 0 12h4zm2 5.291A7.962 7.962 0 014 12H0c0 3.042 1.135 5.824 3 7.938l3-2.647z"></path>
+                    </svg>
+                </span>
+            </div>
+            
+            <!-- Filter by Type -->
+            <div class="mb-4">
+                <h3 class="text-gray-200 font-medium mb-3">Filter by type:</h3>
+                <div class="flex flex-wrap gap-2">
+                    <button hx-post="browse/search/models"
+                        class="inline-flex items-center rounded-full px-4 py-2 text-sm font-medium bg-indigo-900/60 text-indigo-200 border border-indigo-700/50 hover:bg-indigo-800 transition duration-200 ease-in-out"
+                        hx-target="#search-results" 
+                        hx-vals='{"search": "tts"}'
+                        onclick="hidePagination()"
+                        hx-indicator=".htmx-indicator">
+                        <i class="fas fa-microphone mr-2"></i>TTS
+                    </button>
+                    <button hx-post="browse/search/models" 
+                        class="inline-flex items-center rounded-full px-4 py-2 text-sm font-medium bg-purple-900/60 text-purple-200 border border-purple-700/50 hover:bg-purple-800 transition duration-200 ease-in-out"
+                        hx-target="#search-results" 
+                        hx-vals='{"search": "stablediffusion"}'
+                        onclick="hidePagination()"
+                        hx-indicator=".htmx-indicator">
+                        <i class="fas fa-image mr-2"></i>Image generation
+                    </button>
+                    <button hx-post="browse/search/models"
+                        class="inline-flex items-center rounded-full px-4 py-2 text-sm font-medium bg-blue-900/60 text-blue-200 border border-blue-700/50 hover:bg-blue-800 transition duration-200 ease-in-out"
+                        hx-target="#search-results" 
+                        hx-vals='{"search": "llm"}'
+                        onclick="hidePagination()"
+                        hx-indicator=".htmx-indicator">
+                        <i class="fas fa-comment-alt mr-2"></i>Text generation
+                    </button>
+                    <button hx-post="browse/search/models" 
+                        class="inline-flex items-center rounded-full px-4 py-2 text-sm font-medium bg-green-900/60 text-green-200 border border-green-700/50 hover:bg-green-800 transition duration-200 ease-in-out"
+                        hx-target="#search-results" 
+                        hx-vals='{"search": "multimodal"}'
+                        onclick="hidePagination()"
+                        hx-indicator=".htmx-indicator">
+                        <i class="fas fa-object-group mr-2"></i>Multimodal
+                    </button>
+                    <button hx-post="browse/search/models" 
+                        class="inline-flex items-center rounded-full px-4 py-2 text-sm font-medium bg-cyan-900/60 text-cyan-200 border border-cyan-700/50 hover:bg-cyan-800 transition duration-200 ease-in-out"
+                        hx-target="#search-results" 
+                        hx-vals='{"search": "embedding"}'
+                        onclick="hidePagination()"
+                        hx-indicator=".htmx-indicator">
+                        <i class="fas fa-vector-square mr-2"></i>Embeddings
+                    </button>
+                    <button hx-post="browse/search/models"
+                        class="inline-flex items-center rounded-full px-4 py-2 text-sm font-medium bg-amber-900/60 text-amber-200 border border-amber-700/50 hover:bg-amber-800 transition duration-200 ease-in-out"
+                        hx-target="#search-results" 
+                        hx-vals='{"search": "rerank"}'
+                        onclick="hidePagination()"
+                        hx-indicator=".htmx-indicator">
+                        <i class="fas fa-sort-amount-up mr-2"></i>Rerankers
+                    </button>
+                    <button hx-post="browse/search/models"
+                        class="inline-flex items-center rounded-full px-4 py-2 text-sm font-medium bg-teal-900/60 text-teal-200 border border-teal-700/50 hover:bg-teal-800 transition duration-200 ease-in-out"
+                        hx-target="#search-results" 
+                        hx-vals='{"search": "whisper"}'
+                        onclick="hidePagination()"
+                        hx-indicator=".htmx-indicator">
+                        <i class="fas fa-headphones mr-2"></i>Audio transcription
+                    </button>
+                </div>
+            </div>
+            
+            <!-- Filter by Tags -->
+            <div class="mt-5">
+                <h3 class="text-gray-200 font-medium mb-2">Filter by tags:</h3>
+                <div class="flex flex-wrap gap-2 max-h-24 overflow-y-auto scrollbar-thin scrollbar-thumb-gray-700 scrollbar-track-gray-900 pr-2">
+                    {{ range .AllTags }}
+                        <button hx-post="browse/search/models" 
+                            class="inline-flex items-center text-xs px-3 py-1 rounded-full bg-gray-700/60 text-gray-300 border border-gray-600/50 hover:bg-gray-600 hover:text-gray-100 transition duration-200 ease-in-out"
+                            hx-target="#search-results" 
+                            hx-vals='{"search": "{{.}}"}'
+                            onclick="hidePagination()"
+                            hx-indicator=".htmx-indicator">
+                            <i class="fas fa-tag text-xs mr-1.5"></i>{{.}}
+                        </button>
+                    {{ end }}
+                </div>
+            </div>
+        </div>
+
+        <!-- Results Section -->
+        <div id="search-results" class="transition-all duration-300">
+            {{.Models}}
+        </div>
+
+        <!-- Pagination -->
+        {{ if gt .AvailableModels $numModelsPerPage }}
+        <div id="paginate" class="flex justify-center mt-8">
+            <div class="flex items-center gap-4">
+                <button onclick="window.location.href='browse?page={{.PrevPage}}'" 
+                    class="flex items-center justify-center h-10 w-10 bg-gray-800/80 text-gray-300 hover:bg-indigo-900/70 hover:text-white rounded-lg shadow transition duration-300 ease-in-out {{if not .PrevPage}}opacity-50 cursor-not-allowed{{end}}" 
+                    {{if not .PrevPage}}disabled{{end}}>
+                    <i class="fas fa-chevron-left"></i>
+                </button>
+                <div class="text-gray-400 text-sm">
+                    Page <span class="text-white font-medium">{{add .PrevPage 1}}</span>
+                </div>
+                <button onclick="window.location.href='browse?page={{.NextPage}}'" 
+                    class="flex items-center justify-center h-10 w-10 bg-gray-800/80 text-gray-300 hover:bg-indigo-900/70 hover:text-white rounded-lg shadow transition duration-300 ease-in-out {{if not .NextPage}}opacity-50 cursor-not-allowed{{end}}"
+                    {{if not .NextPage}}disabled{{end}}>
+                    <i class="fas fa-chevron-right"></i>
+                </button>
+            </div>
+        </div>
+        {{ end }}
+
+    </div>
     {{template "views/partials/footer" .}}
 </div>
 
+<script>
+    function hidePagination() {
+        const paginateDiv = document.getElementById('paginate');
+        if (paginateDiv) {
+            paginateDiv.style.display = 'none';
+        }
+    }
+
+    // Listen for the htmx:afterSwap event to handle cases when the search results are updated
+    document.body.addEventListener('htmx:afterSwap', function(event) {
+        if (event.detail.target.id === 'search-results') {
+            hidePagination();
+        }
+    });
+</script>
+
 </body>
-</html>
+</html>

core/http/views/p2p.html

@@ -1,172 +1,245 @@
 <!DOCTYPE html>
 <html lang="en">
 {{template "views/partials/head" .}}
-<body class="bg-gray-900 text-gray-200">
+
+<body class="bg-gradient-to-br from-gray-900 to-gray-950 text-gray-200">
 <div class="flex flex-col min-h-screen" x-data="{}">
 
     {{template "views/partials/navbar" .}}
-    <div class="container mx-auto px-4 flex-grow">
-        <div class="workers mt-12 text-center">
-            <div class="animation-container">
+    
+    <div class="container mx-auto px-4 py-8 flex-grow">
+        <div class="workers mt-8">
+            <!-- Hero Section with Network Animation -->
+            <div class="animation-container mb-8">
                 <canvas id="networkCanvas"></canvas>
                 <div class="text-overlay">
-                    <header class="text-center py-12">
-                        <h1 class="text-5xl font-bold text-gray-100">
-                            <i class="fa-solid fa-circle-nodes mr-2"></i> Distributed inference with P2P 
-                        </h1>
-                        <p class="mt-4 text-lg">
-                            Distribute computation by sharing and loadbalancing instances or  sharding model weights.
-                            <a href="https://localai.io/features/distribute/" target="_blank">
-                                <i class="fas fa-circle-info pr-2"></i>
-                            </a>
-                        </p>
-            
-                    </header>
+                    <h1 class="text-4xl md:text-5xl font-bold text-white mb-4">
+                        <span class="bg-clip-text text-transparent bg-gradient-to-r from-blue-400 to-indigo-400">
+                            <i class="fa-solid fa-circle-nodes mr-2"></i> Distributed inference with P2P
+                        </span>
+                    </h1>
+                    <p class="text-xl text-gray-300">
+                        Distribute computation by sharing and loadbalancing instances or sharding model weights
+                        <a href="https://localai.io/features/distribute/" target="_blank" class="text-blue-400 hover:text-blue-300 transition-colors">
+                            <i class="fas fa-circle-info ml-2"></i>
+                        </a>
+                    </p>
                 </div>
-            </div>      
-            <h5 class="mb-4 text-justify">LocalAI uses P2P technologies to enable distribution of work between peers. It is possible to share an instance with Federation and/or split the weights of a model across peers (only available with llama.cpp models). You can now share computational resources between your devices or your friends!</h5>
-
-            <div class="bg-gray-800 p-6 rounded-lg shadow-lg mb-12 text-left">
-                <p class="text-lg font-bold mb-4 mt-1">
-                    Network token
-                    <i class="fa-solid fa-copy copy-icon float-right"></i>
+            </div>
+            
+            <div class="bg-gradient-to-r from-blue-900/30 to-indigo-900/30 rounded-2xl shadow-xl p-6 mb-10">
+                <p class="text-lg mb-4 text-gray-300">
+                    LocalAI uses P2P technologies to enable distribution of work between peers. It is possible to share an instance with Federation and/or split the weights of a model across peers (only available with llama.cpp models). You can now share computational resources between your devices or your friends!
+                </p>
+            </div>
+
+            <!-- Network Token Card -->
+            <div class="bg-gradient-to-r from-gray-800/90 to-gray-800/80 border border-gray-700/50 rounded-xl overflow-hidden shadow-xl mb-10 p-6 transition-all duration-300 hover:shadow-lg hover:shadow-blue-900/20 hover:border-blue-700/50">
+                <div class="flex items-center mb-4">
+                    <i class="fas fa-key text-yellow-400 text-xl mr-3"></i>
+                    <h3 class="text-xl font-bold text-white">Network Token</h3>
+                    <button onclick="copyClipboard('{{.P2PToken}}')" class="ml-auto bg-gray-700 hover:bg-gray-600 text-gray-300 p-2 rounded-lg transition-colors duration-200">
+                        <i class="fa-solid fa-copy"></i>
+                    </button>
+                </div>
+                <code class="block bg-gray-900/80 text-yellow-300 p-4 rounded-lg break-words mb-4 border border-gray-700/50" @click="copyClipboard($el.textContent)">{{.P2PToken}}</code>
+                <p class="text-gray-300">
+                    The network token can be used to either share the instance or join a federation or a worker network. Below you will find examples on how to start a new instance or a worker with this token.
                 </p>
-                <code class="block bg-gray-700 text-yellow-300 p-4 rounded-lg break-words" @click="copyClipboard($el.textContent)">{{.P2PToken}}</code><br>
-                The network token can be used to either share the instance or join a federation or a worker network. Below you will find a few examples on how to start a new instance or a worker with the token and you will be able to see the available workers and federated nodes.
             </div>
         
             <!-- Warning box if p2p token is empty and p2p is enabled -->
             {{ if and .IsP2PEnabled (eq .P2PToken "") }}
-            <div class="bg-red-500 p-4 rounded-lg shadow-lg mb-12 text-left">
-                <p class="text-xl font-semibold text-white"> <i class="fa-solid fa-exclamation-triangle"></i> Warning: P2P mode is disabled or no token was specified</p>
-                <p class="mb-4">You have to enable P2P mode by starting LocalAI with <code>--p2p</code>. Please restart the server with <code>--p2p</code> to generate a new token automatically that can be used to automatically discover other nodes. If you already have a token specify it with <code>export TOKEN=".."</code> <a href="https://localai.io/features/distribute/" target="_blank">
-                    Check out the documentation for more information.
-                </a> </p>
+            <div class="bg-gradient-to-r from-red-800/70 to-red-700/70 border border-red-600/50 p-6 rounded-xl shadow-lg mb-10 text-left">
+                <div class="flex items-center mb-2">
+                    <i class="fa-solid fa-exclamation-triangle text-red-300 text-2xl mr-3"></i>
+                    <h3 class="text-xl font-bold text-white">Warning: P2P mode is disabled or no token was specified</h3>
+                </div>
+                <p class="mb-4 text-red-200">
+                    You have to enable P2P mode by starting LocalAI with <code class="bg-red-900/50 px-2 py-0.5 rounded">--p2p</code>. Please restart the server with <code class="bg-red-900/50 px-2 py-0.5 rounded">--p2p</code> to generate a new token automatically that can be used to discover other nodes. If you already have a token, specify it with <code class="bg-red-900/50 px-2 py-0.5 rounded">export TOKEN=".."</code>
+                    <a href="https://localai.io/features/distribute/" target="_blank" class="text-red-300 hover:text-red-200 underline underline-offset-2">
+                       Check out the documentation for more information.
+                    </a>
+                </p>
             </div>
             {{ else }}
 
             <!-- Federation Box -->
-            <div class="bg-gray-800 p-6 rounded-lg shadow-lg mb-12 text-left">
+            <div class="bg-gradient-to-r from-gray-800/90 to-gray-800/80 border border-gray-700/50 rounded-xl overflow-hidden shadow-xl mb-10 transition-all duration-300 hover:shadow-lg hover:shadow-blue-900/20">
+                <div class="p-6 border-b border-gray-700/50">
+                    <div class="flex items-center mb-3">
+                        <i class="text-blue-400 fa-solid fa-circle-nodes text-2xl mr-3 fa-spin-pulse"></i>
+                        <h2 class="text-2xl font-bold text-white">Federated Nodes: 
+                            <span class="text-blue-400" hx-get="p2p/ui/workers-federation-stats" hx-trigger="every 1s"></span>
+                        </h2>
+                    </div>
+                    <p class="text-gray-300 mb-4">
+                        You can start LocalAI in federated mode to share your instance, or start the federated server to balance requests between nodes of the federation.
+                    </p>
 
-                <p class="text-xl font-semibold text-gray-200"> <i class="text-gray-200 fa-solid fa-circle-nodes"></i> Federated Nodes: <span hx-get="p2p/ui/workers-federation-stats" hx-trigger="every 1s"></span> </p>
-                <p class="mb-4">You can start LocalAI in federated mode to share your instance, or start the federated server to balance requests between nodes of the federation.</p>
-
-                <div class="grid grid-cols-1 sm:grid-cols-2 md:grid-cols-3 gap-4 mb-12">
-                    <div hx-get="p2p/ui/workers-federation" hx-trigger="every 1s"></div>
+                    <div class="grid grid-cols-1 sm:grid-cols-2 lg:grid-cols-3 gap-4 mb-8">
+                        <div hx-get="p2p/ui/workers-federation" hx-trigger="every 1s"></div>
+                    </div>
                 </div>
 
-                <hr class="border-gray-700 mb-12">
+                <div class="p-6">
+                    <h3 class="text-2xl font-bold text-white mb-6">
+                        <i class="fa-solid fa-book text-blue-400 mr-2"></i> Start a federated instance
+                    </h3>
 
-                <h3 class="text-2xl font-semibold text-gray-100 mb-6"><i class="fa-solid fa-book"></i> Start a federated instance</h3>
+                    <!-- Tabs navigation -->
+                    <ul class="mb-5 flex list-none flex-row flex-wrap ps-0 border border-gray-700/50 rounded-lg overflow-hidden" role="tablist" data-twe-nav-ref>
+                        <li role="presentation" class="flex-auto text-center">
+                            <a href="#tabs-federated-cli" class="tablink block border-0 bg-gray-800 px-7 py-4 text-sm font-medium uppercase leading-tight text-white hover:bg-gray-700 focus:bg-gray-700 data-[twe-nav-active]:border-blue-500 data-[twe-nav-active]:text-blue-400 data-[twe-nav-active]:bg-gray-700 active transition-all duration-200" data-twe-toggle="pill" data-twe-target="#tabs-federated-cli" data-twe-nav-active role="tab" aria-controls="tabs-federated-cli" aria-selected="true">
+                                <i class="fa-solid fa-terminal mr-2"></i> CLI
+                            </a>
+                        </li>
+                        <li role="presentation" class="flex-auto text-center">
+                            <a href="#tabs-federated-docker" class="tablink block border-0 bg-gray-800 px-7 py-4 text-sm font-medium uppercase leading-tight text-white hover:bg-gray-700 focus:bg-gray-700 data-[twe-nav-active]:border-blue-500 data-[twe-nav-active]:text-blue-400 data-[twe-nav-active]:bg-gray-700 transition-all duration-200" data-twe-toggle="pill" data-twe-target="#tabs-federated-docker" role="tab" aria-controls="tabs-federated-docker" aria-selected="false">
+                                <i class="fa-solid fa-box-open mr-2"></i> Container images
+                            </a>
+                        </li>
+                    </ul>
 
-                <!-- Tabs navigation -->
-                <ul class="mb-5 flex list-none flex-row flex-wrap ps-0" role="tablist" data-twe-nav-ref>
-                    <li role="presentation" class="flex-auto text-center">
-                        <a href="#tabs-federated-cli" class="tablink my-2 block border-0 bg-gray-800 px-7 pb-3.5 pt-4 text-xs font-medium uppercase leading-tight text-white hover:bg-gray-700 focus:bg-gray-700 data-[twe-nav-active]:border-yellow-500 data-[twe-nav-active]:text-yellow-500 data-[twe-nav-active]:bg-gray-700 active" data-twe-toggle="pill" data-twe-target="#tabs-federated-cli" data-twe-nav-active role="tab" aria-controls="tabs-federated-cli" aria-selected="true"><i class="fa-solid fa-terminal"></i> CLI</a>
-                    </li>
-                    <li role="presentation" class="flex-auto text-center">
-                        <a href="#tabs-federated-docker" class="tablink my-2 block border-0 bg-gray-800 px-7 pb-3.5 pt-4 text-xs font-medium uppercase leading-tight text-white hover:bg-gray-700 focus:bg-gray-700 data-[twe-nav-active]:border-yellow-500 data-[twe-nav-active]:text-yellow-500 data-[twe-nav-active]:bg-gray-700" data-twe-toggle="pill" data-twe-target="#tabs-federated-docker" role="tab" aria-controls="tabs-federated-docker" aria-selected="false"><i class="fa-solid fa-box-open"></i> Container images</a>
-                    </li>
-                </ul>
+                    <!-- Tabs content -->
+                    <div class="mb-6">                    
+                        <div class="tabcontent hidden opacity-100 transition-opacity duration-150 ease-linear data-[twe-tab-active]:block p-4" id="tabs-federated-cli" role="tabpanel" aria-labelledby="tabs-federated-cli" data-twe-tab-active>
+                            <div class="bg-gray-900/50 rounded-xl border border-gray-700/50 p-6">
+                                <div class="flex items-center justify-between mb-4">
+                                    <h4 class="text-lg font-bold text-white">
+                                        Start a new instance to share:
+                                    </h4>
+                                    <button onclick="copyClipboard('export TOKEN=\'{{.P2PToken}}\'\nlocal-ai run --federated --p2p')" class="bg-gray-700 hover:bg-gray-600 text-gray-300 p-2 rounded-lg transition-colors duration-200">
+                                        <i class="fa-solid fa-copy"></i>
+                                    </button>
+                                </div>
+                                <code class="block bg-gray-800 text-yellow-300 p-4 rounded-lg break-words mb-4 border border-gray-700/50">
+# Start a new instance to share with --federated and a TOKEN<br>
+export TOKEN="<span class="token">{{.P2PToken}}</span>"<br>
+local-ai run --federated --p2p</code>
 
-                <!-- Tabs content -->
-                <div class="mb-6">                    
-                    <div class="tabcontent hidden opacity-100 transition-opacity duration-150 ease-linear data-[twe-tab-active]:block p-4" id="tabs-federated-cli" role="tabpanel" aria-labelledby="tabs-federated-cli" data-twe-tab-active>
+                                <p class="text-gray-400 text-sm mt-2">Note: If you don't have a token do not specify it and use the generated one that you can find in this page.</p>
+                                
+                                <div class="flex items-center justify-between mb-4 mt-8">
+                                    <h4 class="text-lg font-bold text-white">
+                                        Start a new federated load balancer:
+                                    </h4>
+                                    <button onclick="copyClipboard('export TOKEN=\'{{.P2PToken}}\'\nlocal-ai federated')" class="bg-gray-700 hover:bg-gray-600 text-gray-300 p-2 rounded-lg transition-colors duration-200">
+                                        <i class="fa-solid fa-copy"></i>
+                                    </button>
+                                </div>
+                                <code class="block bg-gray-800 text-yellow-300 p-4 rounded-lg break-words mb-4 border border-gray-700/50">
+export TOKEN="<span class="token">{{.P2PToken}}</span>"<br>
+local-ai federated</code>
 
-                        <p class="text-lg font-bold mb-4 mt-1">
-                            To start a new instance to share:
-                            <i class="fa-solid fa-copy copy-icon float-right"></i>
-                        </p>
-                        <code class="block bg-gray-700 text-yellow-300 p-4 rounded-lg break-words" @click="copyClipboard($el.textContent)">
-                            # Start a new instance to share with --federated and a TOKEN<br>
-                            export TOKEN="<span class="token">{{.P2PToken}}</span>"<br>
-                            local-ai run --federated --p2p
-                        </code>
+                                <p class="text-gray-400 text-sm mt-2">Note: Token is needed when starting the federated server.</p>
 
-                        <p class="mt-2">Note: If you don't have a token do not specify it and use the generated one that you can find in this page.</p>
-                        <p class="text-lg font-bold mb-4 mt-1">
-                            To start a new federated load balancer:
-                            <i class="fa-solid fa-copy copy-icon float-right"></i>
-                        </p>
-                        <code class="block bg-gray-700 text-yellow-300 p-4 rounded-lg break-words" @click="copyClipboard($el.textContent)">
-                            export TOKEN="<span class="token">{{.P2PToken}}</span>"<br>
-                            local-ai federated
-                        </code>
+                                <p class="text-gray-300 mt-4">For all the options available, please refer to the <a href="https://localai.io/features/distribute/#starting-workers" target="_blank" class="text-blue-400 hover:text-blue-300 transition-colors">documentation</a>.</p>
+                            </div>
+                        </div>
+                        <div class="tabcontent hidden opacity-0 transition-opacity duration-150 ease-linear data-[twe-tab-active]:block p-4" id="tabs-federated-docker" role="tabpanel" aria-labelledby="tabs-federated-docker">
+                            <div class="bg-gray-900/50 rounded-xl border border-gray-700/50 p-6">
+                                <div class="flex items-center justify-between mb-4">
+                                    <h4 class="text-lg font-bold text-white">
+                                        Start a new federated instance:
+                                    </h4>
+                                    <button onclick="copyClipboard('docker run -ti --net host -e TOKEN=\'{{.P2PToken}}\' --name local-ai -p 8080:8080 localai/localai:latest-cpu run --federated --p2p')" class="bg-gray-700 hover:bg-gray-600 text-gray-300 p-2 rounded-lg transition-colors duration-200">
+                                        <i class="fa-solid fa-copy"></i>
+                                    </button>
+                                </div>
+                                <code class="block bg-gray-800 text-yellow-300 p-4 rounded-lg break-words mb-4 border border-gray-700/50">
+docker run -ti --net host -e TOKEN="<span class="token">{{.P2PToken}}</span>" --name local-ai -p 8080:8080 localai/localai:latest-cpu run --federated --p2p</code>
 
-                        <p class="mt-2">Note: Token is needed when starting the federated server.</p>
+                                <div class="flex items-center justify-between mb-4 mt-8">
+                                    <h4 class="text-lg font-bold text-white">
+                                        Start a new federated server with Docker (port to 9090):
+                                    </h4>
+                                    <button onclick="copyClipboard('docker run -ti --net host -e TOKEN=\'{{.P2PToken}}\' --name local-ai -p 9090:8080 localai/localai:latest-cpu federated')" class="bg-gray-700 hover:bg-gray-600 text-gray-300 p-2 rounded-lg transition-colors duration-200">
+                                        <i class="fa-solid fa-copy"></i>
+                                    </button>
+                                </div>
+                                <code class="block bg-gray-800 text-yellow-300 p-4 rounded-lg break-words mb-4 border border-gray-700/50">
+docker run -ti --net host -e TOKEN="<span class="token">{{.P2PToken}}</span>" --name local-ai -p 9090:8080 localai/localai:latest-cpu federated</code>
 
-                        <p class="mt-2">For all the options available, please refer to the <a href="https://localai.io/features/distribute/#starting-workers" target="_blank" class="text-yellow-300 hover:text-yellow-400">documentation</a>.</p>
-                    </div>
-                    <div class="tabcontent hidden opacity-0 transition-opacity duration-150 ease-linear data-[twe-tab-active]:block p-4" id="tabs-federated-docker" role="tabpanel" aria-labelledby="tabs-federated-docker">
-                        <p class="text-lg font-bold mb-4 mt-1">
-                            To start a new federated instance:
-                            <i class="fa-solid fa-copy copy-icon float-right"></i>
-                        </p>
-                        <code class="block bg-gray-700 text-yellow-300 p-4 rounded-lg break-words" @click="copyClipboard($el.textContent)">
-                            docker run -ti --net host -e TOKEN="<span class="token">{{.P2PToken}}</span>" --name local-ai -p 8080:8080 localai/localai:latest-cpu run --federated --p2p
-                        </code>
-
-                        <p class="text-lg font-bold mb-4 mt-1">
-                            To start a new federated server with Docker (port to 9090):
-                            <i class="fa-solid fa-copy copy-icon float-right"></i>
-                        </p>
-                        <code class="block bg-gray-700 text-yellow-300 p-4 rounded-lg break-words" @click="copyClipboard($el.textContent)">
-                            docker run -ti --net host -e TOKEN="<span class="token">{{.P2PToken}}</span>" --name local-ai -p 9090:8080 localai/localai:latest-cpu federated
-                        </code>
-
-                        <p class="mt-2">For all the options available and see what image to use, please refer to the <a href="https://localai.io/basics/container/" target="_blank" class="text-yellow-300 hover:text-yellow-400">Container images documentation</a> and <a href="https://localai.io/advanced/#cli-parameters" target="_blank" class="text-yellow-300 hover:text-yellow-400">CLI parameters documentation</a>.</p>
+                                <p class="text-gray-300 mt-4">For all the options available and see what image to use, please refer to the <a href="https://localai.io/basics/container/" target="_blank" class="text-blue-400 hover:text-blue-300 transition-colors">Container images documentation</a> and <a href="https://localai.io/advanced/#cli-parameters" target="_blank" class="text-blue-400 hover:text-blue-300 transition-colors">CLI parameters documentation</a>.</p>
+                            </div>
+                        </div>
                     </div>
                 </div>
             </div>
 
             <!-- Llama.cpp Box -->
-
-            <div class="bg-gray-800 p-6 rounded-lg shadow-lg mb-12 text-left">
-
-                <p class="text-xl font-semibold text-gray-200"> <i class="text-gray-200 fa-solid fa-circle-nodes"></i> Workers (llama.cpp): <span hx-get="p2p/ui/workers-stats" hx-trigger="every 1s"></span> </p>
-                <p class="mb-4">You can start llama.cpp workers to distribute weights between the workers and offload part of the computation. To start a new worker, you can use the CLI or Docker.</p>
-
-                <div class="grid grid-cols-1 sm:grid-cols-2 md:grid-cols-3 gap-4 mb-12">
-                    <div hx-get="p2p/ui/workers" hx-trigger="every 1s"></div>
-                </div>
-                <hr class="border-gray-700 mb-12">
-
-                <h3 class="text-2xl font-semibold text-gray-100 mb-6"><i class="fa-solid fa-book"></i> Start a new llama.cpp P2P worker</h3>
-                
-                <!-- Tabs navigation -->
-                <ul class="mb-5 flex list-none flex-row flex-wrap ps-0" role="tablist" data-twe-nav-ref>
-                    <li role="presentation" class="flex-auto text-center">
-                        <a href="#tabs-cli" class="tablink my-2 block border-0 bg-gray-800 px-7 pb-3.5 pt-4 text-xs font-medium uppercase leading-tight text-white hover:bg-gray-700 focus:bg-gray-700 data-[twe-nav-active]:border-yellow-500 data-[twe-nav-active]:text-yellow-500 data-[twe-nav-active]:bg-gray-700 active" data-twe-toggle="pill" data-twe-target="#tabs-cli" data-twe-nav-active role="tab" aria-controls="tabs-cli" aria-selected="true"><i class="fa-solid fa-terminal"></i> CLI</a>
-                    </li>
-                    <li role="presentation" class="flex-auto text-center">
-                        <a href="#tabs-docker" class="tablink my-2 block border-0 bg-gray-800 px-7 pb-3.5 pt-4 text-xs font-medium uppercase leading-tight text-white hover:bg-gray-700 focus:bg-gray-700 data-[twe-nav-active]:border-yellow-500 data-[twe-nav-active]:text-yellow-500 data-[twe-nav-active]:bg-gray-700" data-twe-toggle="pill" data-twe-target="#tabs-docker" role="tab" aria-controls="tabs-docker" aria-selected="false"><i class="fa-solid fa-box-open"></i> Container images</a>
-                    </li>
-                </ul>
-
-                <!-- Tabs content -->
-                <div class="mb-6">
-                    <div class="tabcontent hidden opacity-100 transition-opacity duration-150 ease-linear data-[twe-tab-active]:block p-4" id="tabs-cli" role="tabpanel" aria-labelledby="tabs-cli" data-twe-tab-active>
-                        <p class="text-lg font-bold mb-4 mt-1">
-                            To start a new worker, run the following command:
-                            <i class="fa-solid fa-copy copy-icon float-right"></i>
-                        </p>
-                        <code class="block bg-gray-700 text-yellow-300 p-4 rounded-lg break-words" @click="copyClipboard($el.textContent)">
-                            export TOKEN="<span class="token">{{.P2PToken}}</span>"<br>
-                            local-ai worker p2p-llama-cpp-rpc
-                        </code>
-
-                        <p class="mt-2">For all the options available, please refer to the <a href="https://localai.io/features/distribute/#starting-workers" target="_blank" class="text-yellow-300 hover:text-yellow-400">documentation</a>.</p>
+            <div class="bg-gradient-to-r from-gray-800/90 to-gray-800/80 border border-gray-700/50 rounded-xl overflow-hidden shadow-xl mb-10 transition-all duration-300 hover:shadow-lg hover:shadow-blue-900/20">
+                <div class="p-6 border-b border-gray-700/50">
+                    <div class="flex items-center mb-3">
+                        <i class="text-indigo-400 fa-solid fa-circle-nodes text-2xl mr-3 fa-spin-pulse"></i>
+                        <h2 class="text-2xl font-bold text-white">Workers (llama.cpp): 
+                            <span class="text-indigo-400" hx-get="p2p/ui/workers-stats" hx-trigger="every 1s"></span>
+                        </h2>
                     </div>
-                    <div class="tabcontent hidden opacity-0 transition-opacity duration-150 ease-linear data-[twe-tab-active]:block p-4" id="tabs-docker" role="tabpanel" aria-labelledby="tabs-docker">
-                        <p class="text-lg font-bold mb-4 mt-1">
-                            To start a new worker with docker, run the following command:
-                            <i class="fa-solid fa-copy copy-icon float-right"></i>
-                        </p>
-                        <code class="block bg-gray-700 text-yellow-300 p-4 rounded-lg break-words"  @click="copyClipboard($el.textContent)">
-                            docker run -ti --net host -e TOKEN="<span class="token">{{.P2PToken}}</span>" --name local-ai -p 8080:8080 localai/localai:latest-cpu worker p2p-llama-cpp-rpc
-                        </code>
+                    <p class="text-gray-300 mb-4">
+                        You can start llama.cpp workers to distribute weights between the workers and offload part of the computation. To start a new worker, you can use the CLI or Docker.
+                    </p>
 
-                        <p class="mt-2">For all the options available and see what image to use, please refer to the <a href="https://localai.io/basics/container/" target="_blank" class="text-yellow-300 hover:text-yellow-400">Container images documentation</a> and <a href="https://localai.io/advanced/#cli-parameters" target="_blank" class="text-yellow-300 hover:text-yellow-400">CLI parameters documentation</a>.</p>
+                    <div class="grid grid-cols-1 sm:grid-cols-2 lg:grid-cols-3 gap-4 mb-8">
+                        <div hx-get="p2p/ui/workers" hx-trigger="every 1s"></div>
+                    </div>
+                </div>
+
+                <div class="p-6">
+                    <h3 class="text-2xl font-bold text-white mb-6">
+                        <i class="fa-solid fa-book text-indigo-400 mr-2"></i> Start a new llama.cpp P2P worker
+                    </h3>
+                    
+                    <!-- Tabs navigation -->
+                    <ul class="mb-5 flex list-none flex-row flex-wrap ps-0 border border-gray-700/50 rounded-lg overflow-hidden" role="tablist" data-twe-nav-ref>
+                        <li role="presentation" class="flex-auto text-center">
+                            <a href="#tabs-cli" class="tablink block border-0 bg-gray-800 px-7 py-4 text-sm font-medium uppercase leading-tight text-white hover:bg-gray-700 focus:bg-gray-700 data-[twe-nav-active]:border-indigo-500 data-[twe-nav-active]:text-indigo-400 data-[twe-nav-active]:bg-gray-700 active transition-all duration-200" data-twe-toggle="pill" data-twe-target="#tabs-cli" data-twe-nav-active role="tab" aria-controls="tabs-cli" aria-selected="true">
+                                <i class="fa-solid fa-terminal mr-2"></i> CLI
+                            </a>
+                        </li>
+                        <li role="presentation" class="flex-auto text-center">
+                            <a href="#tabs-docker" class="tablink block border-0 bg-gray-800 px-7 py-4 text-sm font-medium uppercase leading-tight text-white hover:bg-gray-700 focus:bg-gray-700 data-[twe-nav-active]:border-indigo-500 data-[twe-nav-active]:text-indigo-400 data-[twe-nav-active]:bg-gray-700 transition-all duration-200" data-twe-toggle="pill" data-twe-target="#tabs-docker" role="tab" aria-controls="tabs-docker" aria-selected="false">
+                                <i class="fa-solid fa-box-open mr-2"></i> Container images
+                            </a>
+                        </li>
+                    </ul>
+
+                    <!-- Tabs content -->
+                    <div class="mb-6">
+                        <div class="tabcontent hidden opacity-100 transition-opacity duration-150 ease-linear data-[twe-tab-active]:block p-4" id="tabs-cli" role="tabpanel" aria-labelledby="tabs-cli" data-twe-tab-active>
+                            <div class="bg-gray-900/50 rounded-xl border border-gray-700/50 p-6">
+                                <div class="flex items-center justify-between mb-4">
+                                    <h4 class="text-lg font-bold text-white">
+                                        Start a new worker:
+                                    </h4>
+                                    <button onclick="copyClipboard('export TOKEN=\'{{.P2PToken}}\'\nlocal-ai worker p2p-llama-cpp-rpc')" class="bg-gray-700 hover:bg-gray-600 text-gray-300 p-2 rounded-lg transition-colors duration-200">
+                                        <i class="fa-solid fa-copy"></i>
+                                    </button>
+                                </div>
+                                <code class="block bg-gray-800 text-yellow-300 p-4 rounded-lg break-words mb-4 border border-gray-700/50">
+export TOKEN="<span class="token">{{.P2PToken}}</span>"<br>
+local-ai worker p2p-llama-cpp-rpc</code>
+
+                                <p class="text-gray-300 mt-4">For all the options available, please refer to the <a href="https://localai.io/features/distribute/#starting-workers" target="_blank" class="text-indigo-400 hover:text-indigo-300 transition-colors">documentation</a>.</p>
+                            </div>
+                        </div>
+                        <div class="tabcontent hidden opacity-0 transition-opacity duration-150 ease-linear data-[twe-tab-active]:block p-4" id="tabs-docker" role="tabpanel" aria-labelledby="tabs-docker">
+                            <div class="bg-gray-900/50 rounded-xl border border-gray-700/50 p-6">
+                                <div class="flex items-center justify-between mb-4">
+                                    <h4 class="text-lg font-bold text-white">
+                                        Start a new worker with Docker:
+                                    </h4>
+                                    <button onclick="copyClipboard('docker run -ti --net host -e TOKEN=\'{{.P2PToken}}\' --name local-ai -p 8080:8080 localai/localai:latest-cpu worker p2p-llama-cpp-rpc')" class="bg-gray-700 hover:bg-gray-600 text-gray-300 p-2 rounded-lg transition-colors duration-200">
+                                        <i class="fa-solid fa-copy"></i>
+                                    </button>
+                                </div>
+                                <code class="block bg-gray-800 text-yellow-300 p-4 rounded-lg break-words mb-4 border border-gray-700/50">
+docker run -ti --net host -e TOKEN="<span class="token">{{.P2PToken}}</span>" --name local-ai -p 8080:8080 localai/localai:latest-cpu worker p2p-llama-cpp-rpc</code>
+
+                                <p class="text-gray-300 mt-4">For all the options available and see what image to use, please refer to the <a href="https://localai.io/basics/container/" target="_blank" class="text-indigo-400 hover:text-indigo-300 transition-colors">Container images documentation</a> and <a href="https://localai.io/advanced/#cli-parameters" target="_blank" class="text-indigo-400 hover:text-indigo-300 transition-colors">CLI parameters documentation</a>.</p>
+                            </div>
+                        </div>
                     </div>
                 </div>
             </div>
@@ -178,6 +251,7 @@
     {{template "views/partials/footer" .}}
 </div>
 <script src="static/p2panimation.js"></script>
+
 <style>
     .token {
         word-break: break-all;
@@ -187,7 +261,19 @@
         flex-direction: column;
         justify-content: space-between;
     }
+    .fa-circle-nodes {
+        animation: pulseGlow 2s ease-in-out infinite;
+    }
+    @keyframes pulseGlow {
+        0%, 100% { filter: drop-shadow(0 0 2px rgba(96, 165, 250, 0.3)); }
+        50% { filter: drop-shadow(0 0 8px rgba(96, 165, 250, 0.7)); }
+    }
+    .copy-icon:hover, button:hover .fa-copy {
+        color: #60a5fa;
+        transform: scale(1.1);
+        transition: all 0.2s ease;
+    }
 </style>
 
 </body>
-</html>
+</html>