⬆️ Update ggerganov/whisper.cpp (#398)

Signed-off-by: GitHub <noreply@github.com>
Co-authored-by: mudler <mudler@users.noreply.github.com>

.github/release.yml

@@ -0,0 +1,24 @@
+# .github/release.yml
+
+changelog:
+  exclude:
+    labels:
+      - ignore-for-release
+  categories:
+    - title: Breaking Changes 🛠
+      labels:
+        - Semver-Major
+        - breaking-change
+    - title: "Bug fixes :bug:"
+      labels:
+        - bug
+    - title: Exciting New Features 🎉
+      labels:
+        - Semver-Minor
+        - enhancement
+    - title: 👒 Dependencies
+      labels:
+        - dependencies
+    - title: Other Changes
+      labels:
+        - "*"

.github/workflows/bump_deps.yaml

@@ -12,8 +12,8 @@ jobs:
           - repository: "go-skynet/go-llama.cpp"
             variable: "GOLLAMA_VERSION"
             branch: "master"
-          - repository: "go-skynet/go-gpt2.cpp"
-            variable: "GOGPT2_VERSION"
+          - repository: "go-skynet/go-ggml-transformers.cpp"
+            variable: "GOGGMLTRANSFORMERS_VERSION"
             branch: "master"
           - repository: "donomii/go-rwkv.cpp"
             variable: "RWKV_VERSION"

.github/workflows/release.yaml

@@ -0,0 +1,84 @@
+name: Build and Release
+
+on: push
+
+permissions:
+  contents: write
+
+jobs:
+  build-linux:
+    strategy:
+      matrix:
+        include:
+          - build: 'avx2'
+            defines: ''
+          - build: 'avx'
+            defines: '-DLLAMA_AVX2=OFF'
+          - build: 'avx512'
+            defines: '-DLLAMA_AVX512=ON'
+    runs-on: ubuntu-latest
+    steps:
+      - name: Clone
+        uses: actions/checkout@v3
+        with: 
+          submodules: true
+      - name: Dependencies
+        run: |
+          sudo apt-get update
+          sudo apt-get install build-essential ffmpeg
+      - name: Build
+        id: build
+        env:
+          CMAKE_ARGS: "${{ matrix.define }}"
+          BUILD_ID: "${{ matrix.build }}"
+        run: |
+          make dist
+      - uses: actions/upload-artifact@v3
+        with:
+          name: ${{ matrix.build }}
+          path: release/
+      - name: Release
+        uses: softprops/action-gh-release@v1
+        if: startsWith(github.ref, 'refs/tags/')
+        with:
+          files: |
+            release/*
+
+  build-macOS:
+    strategy:
+      matrix:
+        include:
+          - build: 'avx2'
+            defines: ''
+          - build: 'avx'
+            defines: '-DLLAMA_AVX2=OFF'
+          - build: 'avx512'
+            defines: '-DLLAMA_AVX512=ON'
+    runs-on: macOS-latest
+    steps:
+      - name: Clone
+        uses: actions/checkout@v3
+        with: 
+          submodules: true
+
+      - name: Dependencies
+        run: |
+          brew update
+          brew install sdl2 ffmpeg
+      - name: Build
+        id: build
+        env:
+          CMAKE_ARGS: "${{ matrix.define }}"
+          BUILD_ID: "${{ matrix.build }}"
+        run: |
+          make dist
+      - uses: actions/upload-artifact@v3
+        with:
+          name: ${{ matrix.build }}
+          path: release/
+      - name: Release
+        uses: softprops/action-gh-release@v1
+        if: startsWith(github.ref, 'refs/tags/')
+        with:
+          files: |
+            release/*

.github/workflows/release.yml.disabled

@@ -1,26 +0,0 @@
-name: goreleaser
-
-on:
-  push:
-    tags:
-      - 'v*'
-
-jobs:
-  goreleaser:
-    runs-on: ubuntu-latest
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v3
-        with:
-          fetch-depth: 0
-      - name: Set up Go
-        uses: actions/setup-go@v3
-        with:
-          go-version: 1.18
-      - name: Run GoReleaser
-        uses: goreleaser/goreleaser-action@v4
-        with:
-          version: latest
-          args: release --clean
-        env:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

.gitignore

@@ -16,5 +16,7 @@ local-ai
 models/*
 test-models/
 
+release/
+
 # just in case
 .DS_Store

.goreleaser.yaml

@@ -1,15 +0,0 @@
-# Make sure to check the documentation at http://goreleaser.com
-project_name: local-ai
-builds:
-  - ldflags:
-      - -w -s
-    env:
-      - CGO_ENABLED=0
-    goos:
-      - linux
-      - darwin
-      - windows
-    goarch:
-      - amd64
-      - arm64
-    binary: '{{ .ProjectName }}'

Dockerfile

@@ -3,9 +3,13 @@ ARG BUILD_TYPE=
 FROM golang:$GO_VERSION
 ENV REBUILD=true
 WORKDIR /build
-RUN apt-get update && apt-get install -y cmake libgomp1 libopenblas-dev libopenblas-base libopencv-dev libopencv-core-dev libopencv-core4.5 ca-certificates
+RUN apt-get update && apt-get install -y cmake curl libgomp1 libopenblas-dev libopenblas-base libopencv-dev libopencv-core-dev libopencv-core4.5 ca-certificates
 COPY . .
 RUN ln -s /usr/include/opencv4/opencv2/ /usr/include/opencv2
 RUN make build
+ENV HEALTHCHECK_ENDPOINT=http://localhost:8080/readyz
+# Define the health check command
+HEALTHCHECK --interval=30s --timeout=360s --retries=10 \
+  CMD curl -f $HEALTHCHECK_ENDPOINT || exit 1
 EXPOSE 8080
 ENTRYPOINT [ "/build/entrypoint.sh" ]

Dockerfile.dev

@@ -11,6 +11,10 @@ RUN make build
 
 FROM debian:$DEBIAN_VERSION
 COPY --from=builder /build/local-ai /usr/bin/local-ai
-RUN apt-get update && apt-get install -y ca-certificates
+RUN apt-get update && apt-get install -y ca-certificates curl
+ENV HEALTHCHECK_ENDPOINT=http://localhost:8080/readyz
+# Define the health check command
+HEALTHCHECK --interval=30s --timeout=360s --retries=10 \
+  CMD curl -f $HEALTHCHECK_ENDPOINT || exit 1
 EXPOSE 8080
 ENTRYPOINT [ "/usr/bin/local-ai" ]

Makefile

@@ -3,13 +3,13 @@ GOTEST=$(GOCMD) test
 GOVET=$(GOCMD) vet
 BINARY_NAME=local-ai
 
-GOLLAMA_VERSION?=ccf23adfb278c0165d388389a5d60f3fe38e4854
+GOLLAMA_VERSION?=fbec625895ba0c458f783b62c8569135c5e80d79
 GPT4ALL_REPO?=https://github.com/nomic-ai/gpt4all
-GPT4ALL_VERSION?=914519e772fd78c15691dcd0b8bac60d6af514ec
-GOGPT2_VERSION?=7bff56f0224502c1c9ed6258d2a17e8084628827
+GPT4ALL_VERSION?=73db20ba85fbbdc66a56e2619394c0eea40dc72b
+GOGGMLTRANSFORMERS_VERSION?=4f18e5eb75089dc1fc8f1c955bb8f73d18520a46
 RWKV_REPO?=https://github.com/donomii/go-rwkv.cpp
 RWKV_VERSION?=07166da10cb2a9e8854395a4f210464dcea76e47
-WHISPER_CPP_VERSION?=041be06d5881d3c759cc4ed45d655804361237cd
+WHISPER_CPP_VERSION?=9b926844e3ae0ca6a0d13573b2e0349be1a4b573
 BERT_VERSION?=cea1ed76a7f48ef386a8e369f6c82c48cdf2d551
 BLOOMZ_VERSION?=e9366e82abdfe70565644fbfae9651976714efd1
 BUILD_TYPE?=
@@ -17,17 +17,20 @@ CGO_LDFLAGS?=
 CUDA_LIBPATH?=/usr/local/cuda/lib64/
 STABLEDIFFUSION_VERSION?=c0748eca3642d58bcf9521108bcee46959c647dc
 GO_TAGS?=
-
+BUILD_ID?=git
+LD_FLAGS=?=
 OPTIONAL_TARGETS?=
 
+OS := $(shell uname -s)
+ARCH := $(shell uname -m)
 GREEN  := $(shell tput -Txterm setaf 2)
 YELLOW := $(shell tput -Txterm setaf 3)
 WHITE  := $(shell tput -Txterm setaf 7)
 CYAN   := $(shell tput -Txterm setaf 6)
 RESET  := $(shell tput -Txterm sgr0)
 
-C_INCLUDE_PATH=$(shell pwd)/go-llama:$(shell pwd)/go-stable-diffusion/:$(shell pwd)/gpt4all/gpt4all-bindings/golang/:$(shell pwd)/go-gpt2:$(shell pwd)/go-rwkv:$(shell pwd)/whisper.cpp:$(shell pwd)/go-bert:$(shell pwd)/bloomz
-LIBRARY_PATH=$(shell pwd)/go-llama:$(shell pwd)/go-stable-diffusion/:$(shell pwd)/gpt4all/gpt4all-bindings/golang/:$(shell pwd)/go-gpt2:$(shell pwd)/go-rwkv:$(shell pwd)/whisper.cpp:$(shell pwd)/go-bert:$(shell pwd)/bloomz
+C_INCLUDE_PATH=$(shell pwd)/go-llama:$(shell pwd)/go-stable-diffusion/:$(shell pwd)/gpt4all/gpt4all-bindings/golang/:$(shell pwd)/go-ggml-transformers:$(shell pwd)/go-rwkv:$(shell pwd)/whisper.cpp:$(shell pwd)/go-bert:$(shell pwd)/bloomz
+LIBRARY_PATH=$(shell pwd)/go-llama:$(shell pwd)/go-stable-diffusion/:$(shell pwd)/gpt4all/gpt4all-bindings/golang/:$(shell pwd)/go-ggml-transformers:$(shell pwd)/go-rwkv:$(shell pwd)/whisper.cpp:$(shell pwd)/go-bert:$(shell pwd)/bloomz
 
 ifeq ($(BUILD_TYPE),openblas)
 	CGO_LDFLAGS+=-lopenblas
@@ -38,6 +41,11 @@ ifeq ($(BUILD_TYPE),cublas)
 	export LLAMA_CUBLAS=1
 endif
 
+# glibc-static or glibc-devel-static required
+ifeq ($(STATIC),true)
+	LD_FLAGS=-linkmode external -extldflags -static
+endif
+
 ifeq ($(GO_TAGS),stablediffusion)
 	OPTIONAL_TARGETS+=go-stable-diffusion/libstablediffusion.a
 endif
@@ -114,23 +122,23 @@ gpt4all/gpt4all-bindings/golang/libgpt4all.a: gpt4all
 	$(MAKE) -C gpt4all/gpt4all-bindings/golang/ libgpt4all.a
 
 ## CEREBRAS GPT
-go-gpt2: 
-	git clone --recurse-submodules https://github.com/go-skynet/go-gpt2.cpp go-gpt2
-	cd go-gpt2 && git checkout -b build $(GOGPT2_VERSION) && git submodule update --init --recursive --depth 1
+go-ggml-transformers: 
+	git clone --recurse-submodules https://github.com/go-skynet/go-ggml-transformers.cpp go-ggml-transformers
+	cd go-ggml-transformers && git checkout -b build $(GOGPT2_VERSION) && git submodule update --init --recursive --depth 1
 	# This is hackish, but needed as both go-llama and go-gpt4allj have their own version of ggml..
-	@find ./go-gpt2 -type f -name "*.c" -exec sed -i'' -e 's/ggml_/ggml_gpt2_/g' {} +
-	@find ./go-gpt2 -type f -name "*.cpp" -exec sed -i'' -e 's/ggml_/ggml_gpt2_/g' {} +
-	@find ./go-gpt2 -type f -name "*.h" -exec sed -i'' -e 's/ggml_/ggml_gpt2_/g' {} +
-	@find ./go-gpt2 -type f -name "*.cpp" -exec sed -i'' -e 's/gpt_print_usage/gpt2_print_usage/g' {} +
-	@find ./go-gpt2 -type f -name "*.h" -exec sed -i'' -e 's/gpt_print_usage/gpt2_print_usage/g' {} +
-	@find ./go-gpt2 -type f -name "*.cpp" -exec sed -i'' -e 's/gpt_params_parse/gpt2_params_parse/g' {} +
-	@find ./go-gpt2 -type f -name "*.h" -exec sed -i'' -e 's/gpt_params_parse/gpt2_params_parse/g' {} +
-	@find ./go-gpt2 -type f -name "*.cpp" -exec sed -i'' -e 's/gpt_random_prompt/gpt2_random_prompt/g' {} +
-	@find ./go-gpt2 -type f -name "*.h" -exec sed -i'' -e 's/gpt_random_prompt/gpt2_random_prompt/g' {} +
-	@find ./go-gpt2 -type f -name "*.cpp" -exec sed -i'' -e 's/json_/json_gpt2_/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.c" -exec sed -i'' -e 's/ggml_/ggml_gpt2_/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.cpp" -exec sed -i'' -e 's/ggml_/ggml_gpt2_/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.h" -exec sed -i'' -e 's/ggml_/ggml_gpt2_/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.cpp" -exec sed -i'' -e 's/gpt_print_usage/gpt2_print_usage/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.h" -exec sed -i'' -e 's/gpt_print_usage/gpt2_print_usage/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.cpp" -exec sed -i'' -e 's/gpt_params_parse/gpt2_params_parse/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.h" -exec sed -i'' -e 's/gpt_params_parse/gpt2_params_parse/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.cpp" -exec sed -i'' -e 's/gpt_random_prompt/gpt2_random_prompt/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.h" -exec sed -i'' -e 's/gpt_random_prompt/gpt2_random_prompt/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.cpp" -exec sed -i'' -e 's/json_/json_gpt2_/g' {} +
 
-go-gpt2/libgpt2.a: go-gpt2
-	$(MAKE) -C go-gpt2 libgpt2.a
+go-ggml-transformers/libtransformers.a: go-ggml-transformers
+	$(MAKE) -C go-ggml-transformers libtransformers.a
 
 whisper.cpp:
 	git clone https://github.com/ggerganov/whisper.cpp.git
@@ -152,21 +160,21 @@ go-llama/libbinding.a: go-llama
 replace:
 	$(GOCMD) mod edit -replace github.com/go-skynet/go-llama.cpp=$(shell pwd)/go-llama
 	$(GOCMD) mod edit -replace github.com/nomic-ai/gpt4all/gpt4all-bindings/golang=$(shell pwd)/gpt4all/gpt4all-bindings/golang
-	$(GOCMD) mod edit -replace github.com/go-skynet/go-gpt2.cpp=$(shell pwd)/go-gpt2
+	$(GOCMD) mod edit -replace github.com/go-skynet/go-ggml-transformers.cpp=$(shell pwd)/go-ggml-transformers
 	$(GOCMD) mod edit -replace github.com/donomii/go-rwkv.cpp=$(shell pwd)/go-rwkv
 	$(GOCMD) mod edit -replace github.com/ggerganov/whisper.cpp=$(shell pwd)/whisper.cpp
 	$(GOCMD) mod edit -replace github.com/go-skynet/go-bert.cpp=$(shell pwd)/go-bert
 	$(GOCMD) mod edit -replace github.com/go-skynet/bloomz.cpp=$(shell pwd)/bloomz
 	$(GOCMD) mod edit -replace github.com/mudler/go-stable-diffusion=$(shell pwd)/go-stable-diffusion
 
-prepare-sources: go-llama go-gpt2 gpt4all go-rwkv whisper.cpp go-bert bloomz go-stable-diffusion replace
+prepare-sources: go-llama go-ggml-transformers gpt4all go-rwkv whisper.cpp go-bert bloomz go-stable-diffusion replace
 	$(GOCMD) mod download
 
 ## GENERIC
 rebuild: ## Rebuilds the project
 	$(MAKE) -C go-llama clean
 	$(MAKE) -C gpt4all/gpt4all-bindings/golang/ clean
-	$(MAKE) -C go-gpt2 clean
+	$(MAKE) -C go-ggml-transformers clean
 	$(MAKE) -C go-rwkv clean
 	$(MAKE) -C whisper.cpp clean
 	$(MAKE) -C go-stable-diffusion clean
@@ -174,18 +182,19 @@ rebuild: ## Rebuilds the project
 	$(MAKE) -C bloomz clean
 	$(MAKE) build
 
-prepare: prepare-sources gpt4all/gpt4all-bindings/golang/libgpt4all.a $(OPTIONAL_TARGETS) go-llama/libbinding.a go-bert/libgobert.a go-gpt2/libgpt2.a go-rwkv/librwkv.a whisper.cpp/libwhisper.a bloomz/libbloomz.a  ## Prepares for building
+prepare: prepare-sources gpt4all/gpt4all-bindings/golang/libgpt4all.a $(OPTIONAL_TARGETS) go-llama/libbinding.a go-bert/libgobert.a go-ggml-transformers/libtransformers.a go-rwkv/librwkv.a whisper.cpp/libwhisper.a bloomz/libbloomz.a  ## Prepares for building
 
 clean: ## Remove build related file
 	rm -fr ./go-llama
 	rm -rf ./gpt4all
 	rm -rf ./go-stable-diffusion
-	rm -rf ./go-gpt2
+	rm -rf ./go-ggml-transformers
 	rm -rf ./go-rwkv
 	rm -rf ./go-bert
 	rm -rf ./bloomz
 	rm -rf ./whisper.cpp
 	rm -rf $(BINARY_NAME)
+	rm -rf release/
 
 ## Build:
 
@@ -193,7 +202,11 @@ build: prepare ## Build the project
 	$(info ${GREEN}I local-ai build info:${RESET})
 	$(info ${GREEN}I BUILD_TYPE: ${YELLOW}$(BUILD_TYPE)${RESET})
 	$(info ${GREEN}I GO_TAGS: ${YELLOW}$(GO_TAGS)${RESET})
-	CGO_LDFLAGS="$(CGO_LDFLAGS)" C_INCLUDE_PATH=${C_INCLUDE_PATH} LIBRARY_PATH=${LIBRARY_PATH} $(GOCMD) build -tags "$(GO_TAGS)" -x -o $(BINARY_NAME) ./
+	CGO_LDFLAGS="$(CGO_LDFLAGS)" C_INCLUDE_PATH=${C_INCLUDE_PATH} LIBRARY_PATH=${LIBRARY_PATH} $(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -x -o $(BINARY_NAME) ./
+
+dist: build
+	mkdir -p release
+	cp $(BINARY_NAME) release/$(BINARY_NAME)-$(BUILD_ID)-$(OS)-$(ARCH)
 
 generic-build: ## Build the project using generic
 	BUILD_TYPE="generic" $(MAKE) build
@@ -205,7 +218,7 @@ run: prepare ## run local-ai
 test-models/testmodel:
 	mkdir test-models
 	mkdir test-dir
-	wget https://huggingface.co/concedo/cerebras-111M-ggml/resolve/main/cerberas-111m-q4_0.bin -O test-models/testmodel
+	wget https://huggingface.co/nnakasato/ggml-model-test/resolve/main/ggml-model-q4.bin -O test-models/testmodel
 	wget https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-base.en.bin -O test-models/whisper-en
 	wget https://huggingface.co/skeskinen/ggml/resolve/main/all-MiniLM-L6-v2/ggml-model-q4_0.bin -O test-models/bert
 	wget https://cdn.openai.com/whisper/draft-20220913a/micro-machines.wav -O test-dir/audio.wav
@@ -215,7 +228,7 @@ test-models/testmodel:
 
 test: prepare test-models/testmodel
 	cp tests/models_fixtures/* test-models
-	C_INCLUDE_PATH=${C_INCLUDE_PATH} LIBRARY_PATH=${LIBRARY_PATH} TEST_DIR=$(abspath ./)/test-dir/ FIXTURES=$(abspath ./)/tests/fixtures CONFIG_FILE=$(abspath ./)/test-models/config.yaml MODELS_PATH=$(abspath ./)/test-models $(GOCMD) run github.com/onsi/ginkgo/v2/ginkgo -v -r ./api ./pkg
+	C_INCLUDE_PATH=${C_INCLUDE_PATH} LIBRARY_PATH=${LIBRARY_PATH} TEST_DIR=$(abspath ./)/test-dir/ FIXTURES=$(abspath ./)/tests/fixtures CONFIG_FILE=$(abspath ./)/test-models/config.yaml MODELS_PATH=$(abspath ./)/test-models $(GOCMD) run github.com/onsi/ginkgo/v2/ginkgo --flakeAttempts 5 -v -r ./api ./pkg
 
 ## Help:
 help: ## Show this help.

README.md

@@ -23,7 +23,7 @@ In a nutshell:
 
 LocalAI is a community-driven project, focused on making the AI accessible to anyone. Any contribution, feedback and PR is welcome! It was initially created by [mudler](https://github.com/mudler/) at the [SpectroCloud OSS Office](https://github.com/spectrocloud).
 
-See the [usage](https://github.com/go-skynet/LocalAI#usage) and [examples](https://github.com/go-skynet/LocalAI/tree/master/examples/) sections to learn how to use LocalAI.
+See the [usage](https://github.com/go-skynet/LocalAI#usage) and [examples](https://github.com/go-skynet/LocalAI/tree/master/examples/) sections to learn how to use LocalAI. For a list of curated models check out the [model gallery](https://github.com/go-skynet/model-gallery).
 
 ### How does it work?  
 
@@ -39,6 +39,7 @@ LocalAI uses C++ bindings for optimizing speed. It is based on [llama.cpp](https
 
 ## News
 
+- 23-05-2023: __v1.15.0__ released. `go-gpt2.cpp` backend got renamed to `go-ggml-transformers.cpp` updated including https://github.com/ggerganov/llama.cpp/pull/1508 which breaks compatibility with older models. This impacts RedPajama, GptNeoX, MPT(not `gpt4all-mpt`), Dolly, GPT2 and Starcoder based models. [Binary releases available](https://github.com/go-skynet/LocalAI/releases), various fixes, including https://github.com/go-skynet/LocalAI/pull/341 .
 - 21-05-2023: __v1.14.0__ released. Minor updates to the `/models/apply` endpoint, `llama.cpp` backend updated including https://github.com/ggerganov/llama.cpp/pull/1508 which breaks compatibility with older models. `gpt4all` is still compatible with the old format. 
 - 19-05-2023: __v1.13.0__ released! 🔥🔥 updates to the `gpt4all` and `llama` backend, consolidated CUDA support ( https://github.com/go-skynet/LocalAI/pull/310 thanks to @bubthegreat and @Thireus ), preliminar support for [installing models via API](https://github.com/go-skynet/LocalAI#advanced-prepare-models-using-the-api).
 - 17-05-2023:  __v1.12.0__ released! 🔥🔥 Minor fixes, plus CUDA (https://github.com/go-skynet/LocalAI/pull/258) support for `llama.cpp`-compatible models and image generation (https://github.com/go-skynet/LocalAI/pull/272).
@@ -62,8 +63,9 @@ Now LocalAI can generate images too:
 
 Twitter: [@LocalAI_API](https://twitter.com/LocalAI_API) and [@mudler_it](https://twitter.com/mudler_it)
 
-### Blogs and articles
+### Blogs, articles, media
 
+- [LocalAI meets k8sgpt](https://www.youtube.com/watch?v=PKrDNuJ_dfE) - CNCF Webinar showcasing LocalAI and k8sgpt.
 - [Question Answering on Documents locally with LangChain, LocalAI, Chroma, and GPT4All](https://mudler.pm/posts/localai-question-answering/) by Ettore Di Giacinto
 - [Tutorial to use k8sgpt with LocalAI](https://medium.com/@tyler_97636/k8sgpt-localai-unlock-kubernetes-superpowers-for-free-584790de9b65) - excellent usecase for localAI, using AI to analyse Kubernetes clusters. by Tyller Gillson
 
@@ -129,13 +131,13 @@ Depending on the model you are attempting to run might need more RAM or CPU reso
 | [gpt4all-llama](https://github.com/nomic-ai/gpt4all)      | Vicuna, Alpaca, LLaMa | yes                      | no                        | no                                | yes                  |
 | [gpt4all-mpt](https://github.com/nomic-ai/gpt4all)          | MPT                   | yes                      | no                        | no                                | yes                  |
 | [gpt4all-j](https://github.com/nomic-ai/gpt4all)           | GPT4ALL-J             | yes                      | no                        | no                                | yes                  |
-| [gpt2](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-gpt2.cpp))             | GPT/NeoX, Cerebras    | yes                      | no                        | no                                | no                   |
-| [dolly](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-gpt2.cpp))            | Dolly                 | yes                      | no                        | no                                | no                   |
-| [redpajama](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-gpt2.cpp))        | RedPajama             | yes                      | no                        | no                                | no                   |
-| [stableLM](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-gpt2.cpp))         | StableLM GPT/NeoX     | yes                      | no                        | no                                | no                   |
-| [replit](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-gpt2.cpp))        | Replit             | yes                      | no                        | no                                | no                   |
-| [gptneox](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-gpt2.cpp))        | GPT NeoX             | yes                      | no                        | no                                | no                   |
-| [starcoder](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-gpt2.cpp))        | Starcoder             | yes                      | no                        | no                                | no                   |
+| [gpt2](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp))             | GPT2, Cerebras    | yes                      | no                        | no                                | no                   |
+| [dolly](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp))            | Dolly                 | yes                      | no                        | no                                | no                   |
+| [gptj](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp))        | GPTJ             | yes                      | no                        | no                                | no                   |
+| [mpt](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp))         | MPT     | yes                      | no                        | no                                | no                   |
+| [replit](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp))        | Replit             | yes                      | no                        | no                                | no                   |
+| [gptneox](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp))        | GPT NeoX, RedPajama, StableLM             | yes                      | no                        | no                                | no                   |
+| [starcoder](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp))        | Starcoder             | yes                      | no                        | no                                | no                   |
 | [bloomz](https://github.com/NouamaneTazi/bloomz.cpp) ([binding](https://github.com/go-skynet/bloomz.cpp))       | Bloom                 | yes                      | no                        | no                                | no                   |
 | [rwkv](https://github.com/saharNooby/rwkv.cpp) ([binding](https://github.com/donomii/go-rw))       | rwkv                 | yes                      | no                        | no                                | yes                   |
 | [bert](https://github.com/skeskinen/bert.cpp) ([binding](https://github.com/go-skynet/go-bert.cpp) | bert                  | no                       | no                  | yes                               | no                   |    
@@ -607,47 +609,69 @@ It should work, however you need to make sure you give enough resources to the c
 LocalAI can be installed inside Kubernetes with helm.
 
 <details>
+By default, the helm chart will install LocalAI instance using the ggml-gpt4all-j model without persistent storage.
 
 1. Add the helm repo
     ```bash
     helm repo add go-skynet https://go-skynet.github.io/helm-charts/
     ```
-1. Create a values files with your settings:
-```bash
-cat <<EOF > values.yaml
+2. Install the helm chart:
+    ```bash
+    helm repo update
+    helm install local-ai go-skynet/local-ai -f values.yaml
+    ```
+> **Note:** For further configuration options, see the [helm chart repository on GitHub](https://github.com/go-skynet/helm-charts).
+### Example values
+Deploy a single LocalAI pod with 6GB of persistent storage serving up a `ggml-gpt4all-j` model with custom prompt.
+```yaml
+### values.yaml
+
 deployment:
-  image: quay.io/go-skynet/local-ai:latest
+  # Adjust the number of threads and context size for model inference
   env:
-    threads: 4
-    contextSize: 1024
-    modelsPath: "/models"
-# Optionally create a PVC, mount the PV to the LocalAI Deployment,
-# and download a model to prepopulate the models directory
-modelsVolume:
-  enabled: true
-  url: "https://gpt4all.io/models/ggml-gpt4all-j.bin"
-  pvc:
+    threads: 14
+    contextSize: 512
+
+# Set the pod requests/limits
+resources:
+  limits:
+    cpu: 4000m
+    memory: 7000Mi
+  requests:
+    cpu: 100m
+    memory: 6000Mi
+
+# Add a custom prompt template for the ggml-gpt4all-j model
+promptTemplates:
+  # The name of the model this template belongs to
+  ggml-gpt4all-j.bin.tmpl: |
+    This is my custom prompt template...
+    ### Prompt:
+    {{.Input}}
+    ### Response:
+
+# Model configuration
+models:
+  # Don't re-download models on pod creation
+  forceDownload: false
+
+  # List of models to download and serve
+  list:
+    - url: "https://gpt4all.io/models/ggml-gpt4all-j.bin"
+       # Optional basic HTTP authentication
+      basicAuth: base64EncodedCredentials
+  
+  # Enable 6Gb of persistent storage models and prompt templates
+  persistence:
+    enabled: true
     size: 6Gi
-    accessModes:
-    - ReadWriteOnce
-  auth:
-    # Optional value for HTTP basic access authentication header
-    basic: "" # 'username:password' base64 encoded
+
 service:
   type: ClusterIP
   annotations: {}
   # If using an AWS load balancer, you'll need to override the default 60s load balancer idle timeout
   # service.beta.kubernetes.io/aws-load-balancer-connection-idle-timeout: "1200"
-EOF
 ```
-3. Install the helm chart:
-```bash
-helm repo update
-helm install local-ai go-skynet/local-ai -f values.yaml
-```
-
-Check out also the [helm chart repository on GitHub](https://github.com/go-skynet/helm-charts).
-
 </details>
 
 ## Supported OpenAI API endpoints
@@ -1001,7 +1025,7 @@ There is the availability of localai-webui and chatbot-ui in the examples sectio
 
 <details>
 
-AutoGPT currently doesn't allow to set a different API URL, but there is a PR open for it, so this should be possible soon!
+Yes, see the [examples](https://github.com/go-skynet/LocalAI/tree/master/examples/)!
 
 </details>
 
@@ -1012,11 +1036,8 @@ Feel free to open up a PR to get your project listed!
 - [Kairos](https://github.com/kairos-io/kairos)
 - [k8sgpt](https://github.com/k8sgpt-ai/k8sgpt#running-local-models)
 - [Spark](https://github.com/cedriking/spark)
-
-## Blog posts and other articles
-
-- https://medium.com/@tyler_97636/k8sgpt-localai-unlock-kubernetes-superpowers-for-free-584790de9b65
-- https://kairos.io/docs/examples/localai/
+- [autogpt4all](https://github.com/aorumbayev/autogpt4all)
+- [Mods](https://github.com/charmbracelet/mods)
 
 ## Short-term roadmap
 
@@ -1045,7 +1066,7 @@ MIT
 
 - [go-skynet/go-llama.cpp](https://github.com/go-skynet/go-llama.cpp)
 - [go-skynet/go-gpt4all-j.cpp](https://github.com/go-skynet/go-gpt4all-j.cpp)
-- [go-skynet/go-gpt2.cpp](https://github.com/go-skynet/go-gpt2.cpp)
+- [go-skynet/go-ggml-transformers.cpp](https://github.com/go-skynet/go-ggml-transformers.cpp)
 - [go-skynet/go-bert.cpp](https://github.com/go-skynet/go-bert.cpp)
 - [donomii/go-rwkv.cpp](https://github.com/donomii/go-rwkv.cpp)
 

api/api.go

@@ -69,6 +69,18 @@ func App(opts ...AppOption) *fiber.App {
 	// Default middleware config
 	app.Use(recover.New())
 
+	if options.preloadJSONModels != "" {
+		if err := ApplyGalleryFromString(options.loader.ModelPath, options.preloadJSONModels, cm); err != nil {
+			return nil
+		}
+	}
+
+	if options.preloadModelsFromPath != "" {
+		if err := ApplyGalleryFromFile(options.loader.ModelPath, options.preloadModelsFromPath, cm); err != nil {
+			return nil
+		}
+	}
+
 	if options.cors {
 		if options.corsAllowOrigins == "" {
 			app.Use(cors.New())
@@ -114,6 +126,14 @@ func App(opts ...AppOption) *fiber.App {
 		app.Static("/generated-images", options.imageDir)
 	}
 
+	ok := func(c *fiber.Ctx) error {
+		return c.SendStatus(200)
+	}
+
+	// Kubernetes health checks
+	app.Get("/healthz", ok)
+	app.Get("/readyz", ok)
+
 	// models
 	app.Get("/v1/models", listModels(options.loader, cm))
 	app.Get("/models", listModels(options.loader, cm))

api/config.go

@@ -16,24 +16,28 @@ import (
 )
 
 type Config struct {
-	OpenAIRequest               `yaml:"parameters"`
-	Name                        string            `yaml:"name"`
-	StopWords                   []string          `yaml:"stopwords"`
-	Cutstrings                  []string          `yaml:"cutstrings"`
-	TrimSpace                   []string          `yaml:"trimspace"`
-	ContextSize                 int               `yaml:"context_size"`
-	F16                         bool              `yaml:"f16"`
-	Threads                     int               `yaml:"threads"`
-	Debug                       bool              `yaml:"debug"`
-	Roles                       map[string]string `yaml:"roles"`
-	Embeddings                  bool              `yaml:"embeddings"`
-	Backend                     string            `yaml:"backend"`
-	TemplateConfig              TemplateConfig    `yaml:"template"`
-	MirostatETA                 float64           `yaml:"mirostat_eta"`
-	MirostatTAU                 float64           `yaml:"mirostat_tau"`
-	Mirostat                    int               `yaml:"mirostat"`
-	NGPULayers                  int               `yaml:"gpu_layers"`
-	ImageGenerationAssets       string            `yaml:"asset_dir"`
+	OpenAIRequest         `yaml:"parameters"`
+	Name                  string            `yaml:"name"`
+	StopWords             []string          `yaml:"stopwords"`
+	Cutstrings            []string          `yaml:"cutstrings"`
+	TrimSpace             []string          `yaml:"trimspace"`
+	ContextSize           int               `yaml:"context_size"`
+	F16                   bool              `yaml:"f16"`
+	Threads               int               `yaml:"threads"`
+	Debug                 bool              `yaml:"debug"`
+	Roles                 map[string]string `yaml:"roles"`
+	Embeddings            bool              `yaml:"embeddings"`
+	Backend               string            `yaml:"backend"`
+	TemplateConfig        TemplateConfig    `yaml:"template"`
+	MirostatETA           float64           `yaml:"mirostat_eta"`
+	MirostatTAU           float64           `yaml:"mirostat_tau"`
+	Mirostat              int               `yaml:"mirostat"`
+	NGPULayers            int               `yaml:"gpu_layers"`
+	ImageGenerationAssets string            `yaml:"asset_dir"`
+
+	PromptCachePath string `yaml:"prompt_cache_path"`
+	PromptCacheAll  bool   `yaml:"prompt_cache_all"`
+
 	PromptStrings, InputStrings []string
 	InputToken                  [][]int
 }

api/gallery.go

@@ -2,10 +2,12 @@ package api
 
 import (
 	"context"
+	"encoding/json"
 	"fmt"
 	"io/ioutil"
 	"net/http"
 	"net/url"
+	"os"
 	"strings"
 	"sync"
 
@@ -40,6 +42,43 @@ func newGalleryApplier(modelPath string) *galleryApplier {
 		statuses:  make(map[string]*galleryOpStatus),
 	}
 }
+
+func applyGallery(modelPath string, req ApplyGalleryModelRequest, cm *ConfigMerger) error {
+	url, err := req.DecodeURL()
+	if err != nil {
+		return err
+	}
+
+	// Send a GET request to the URL
+	response, err := http.Get(url)
+	if err != nil {
+		return err
+	}
+	defer response.Body.Close()
+
+	// Read the response body
+	body, err := ioutil.ReadAll(response.Body)
+	if err != nil {
+		return err
+	}
+
+	// Unmarshal YAML data into a Config struct
+	var config gallery.Config
+	err = yaml.Unmarshal(body, &config)
+	if err != nil {
+		return err
+	}
+
+	config.Files = append(config.Files, req.AdditionalFiles...)
+
+	if err := gallery.Apply(modelPath, req.Name, &config, req.Overrides); err != nil {
+		return err
+	}
+
+	// Reload models
+	return cm.LoadConfigs(modelPath)
+}
+
 func (g *galleryApplier) updatestatus(s string, op *galleryOpStatus) {
 	g.Lock()
 	defer g.Unlock()
@@ -66,44 +105,7 @@ func (g *galleryApplier) start(c context.Context, cm *ConfigMerger) {
 					g.updatestatus(op.id, &galleryOpStatus{Error: e, Processed: true})
 				}
 
-				url, err := op.req.DecodeURL()
-				if err != nil {
-					updateError(err)
-					continue
-				}
-
-				// Send a GET request to the URL
-				response, err := http.Get(url)
-				if err != nil {
-					updateError(err)
-					continue
-				}
-				defer response.Body.Close()
-
-				// Read the response body
-				body, err := ioutil.ReadAll(response.Body)
-				if err != nil {
-					updateError(err)
-					continue
-				}
-
-				// Unmarshal YAML data into a Config struct
-				var config gallery.Config
-				err = yaml.Unmarshal(body, &config)
-				if err != nil {
-					updateError(fmt.Errorf("failed to unmarshal YAML: %v", err))
-					continue
-				}
-
-				config.Files = append(config.Files, op.req.AdditionalFiles...)
-
-				if err := gallery.Apply(g.modelPath, op.req.Name, &config, op.req.Overrides); err != nil {
-					updateError(err)
-					continue
-				}
-
-				// Reload models
-				if err := cm.LoadConfigs(g.modelPath); err != nil {
+				if err := applyGallery(g.modelPath, op.req, cm); err != nil {
 					updateError(err)
 					continue
 				}
@@ -114,6 +116,41 @@ func (g *galleryApplier) start(c context.Context, cm *ConfigMerger) {
 	}()
 }
 
+func ApplyGalleryFromFile(modelPath, s string, cm *ConfigMerger) error {
+	dat, err := os.ReadFile(s)
+	if err != nil {
+		return err
+	}
+	var requests []ApplyGalleryModelRequest
+	err = json.Unmarshal(dat, &requests)
+	if err != nil {
+		return err
+	}
+
+	for _, r := range requests {
+		if err := applyGallery(modelPath, r, cm); err != nil {
+			return err
+		}
+	}
+
+	return nil
+}
+func ApplyGalleryFromString(modelPath, s string, cm *ConfigMerger) error {
+	var requests []ApplyGalleryModelRequest
+	err := json.Unmarshal([]byte(s), &requests)
+	if err != nil {
+		return err
+	}
+
+	for _, r := range requests {
+		if err := applyGallery(modelPath, r, cm); err != nil {
+			return err
+		}
+	}
+
+	return nil
+}
+
 // endpoints
 
 type ApplyGalleryModelRequest struct {

api/options.go

@@ -15,6 +15,8 @@ type Option struct {
 	debug, disableMessage           bool
 	imageDir                        string
 	cors                            bool
+	preloadJSONModels               string
+	preloadModelsFromPath           string
 	corsAllowOrigins                string
 }
 
@@ -53,6 +55,17 @@ func WithContext(ctx context.Context) AppOption {
 	}
 }
 
+func WithYAMLConfigPreload(configFile string) AppOption {
+	return func(o *Option) {
+		o.preloadModelsFromPath = configFile
+	}
+}
+
+func WithJSONStringPreload(configFile string) AppOption {
+	return func(o *Option) {
+		o.preloadJSONModels = configFile
+	}
+}
 func WithConfigFile(configFile string) AppOption {
 	return func(o *Option) {
 		o.configFile = configFile

api/prediction.go

@@ -2,6 +2,8 @@ package api
 
 import (
 	"fmt"
+	"os"
+	"path/filepath"
 	"regexp"
 	"strings"
 	"sync"
@@ -11,7 +13,7 @@ import (
 	"github.com/go-skynet/LocalAI/pkg/stablediffusion"
 	"github.com/go-skynet/bloomz.cpp"
 	bert "github.com/go-skynet/go-bert.cpp"
-	gpt2 "github.com/go-skynet/go-gpt2.cpp"
+	transformers "github.com/go-skynet/go-ggml-transformers.cpp"
 	llama "github.com/go-skynet/go-llama.cpp"
 	gpt4all "github.com/nomic-ai/gpt4all/gpt4all-bindings/golang"
 )
@@ -102,7 +104,7 @@ func ModelEmbedding(s string, tokens []int, loader *model.ModelLoader, c Config)
 	switch model := inferenceModel.(type) {
 	case *llama.LLama:
 		fn = func() ([]float32, error) {
-			predictOptions := buildLLamaPredictOptions(c)
+			predictOptions := buildLLamaPredictOptions(c, loader.ModelPath)
 			if len(tokens) > 0 {
 				return model.TokenEmbeddings(tokens, predictOptions...)
 			}
@@ -151,7 +153,7 @@ func ModelEmbedding(s string, tokens []int, loader *model.ModelLoader, c Config)
 	}, nil
 }
 
-func buildLLamaPredictOptions(c Config) []llama.PredictOption {
+func buildLLamaPredictOptions(c Config, modelPath string) []llama.PredictOption {
 	// Generate the prediction using the language model
 	predictOptions := []llama.PredictOption{
 		llama.SetTemperature(c.Temperature),
@@ -161,6 +163,17 @@ func buildLLamaPredictOptions(c Config) []llama.PredictOption {
 		llama.SetThreads(c.Threads),
 	}
 
+	if c.PromptCacheAll {
+		predictOptions = append(predictOptions, llama.EnablePromptCacheAll)
+	}
+
+	if c.PromptCachePath != "" {
+		// Create parent directory
+		p := filepath.Join(modelPath, c.PromptCachePath)
+		os.MkdirAll(filepath.Dir(p), 0755)
+		predictOptions = append(predictOptions, llama.SetPathPromptCache(p))
+	}
+
 	if c.Mirostat != 0 {
 		predictOptions = append(predictOptions, llama.SetMirostat(c.Mirostat))
 	}
@@ -243,23 +256,23 @@ func ModelInference(s string, loader *model.ModelLoader, c Config, tokenCallback
 
 			return response, nil
 		}
-	case *gpt2.GPTNeoX:
+	case *transformers.GPTNeoX:
 		fn = func() (string, error) {
 			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
+			predictOptions := []transformers.PredictOption{
+				transformers.SetTemperature(c.Temperature),
+				transformers.SetTopP(c.TopP),
+				transformers.SetTopK(c.TopK),
+				transformers.SetTokens(c.Maxtokens),
+				transformers.SetThreads(c.Threads),
 			}
 
 			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
+				predictOptions = append(predictOptions, transformers.SetBatch(c.Batch))
 			}
 
 			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
+				predictOptions = append(predictOptions, transformers.SetSeed(c.Seed))
 			}
 
 			return model.Predict(
@@ -267,23 +280,23 @@ func ModelInference(s string, loader *model.ModelLoader, c Config, tokenCallback
 				predictOptions...,
 			)
 		}
-	case *gpt2.Replit:
+	case *transformers.Replit:
 		fn = func() (string, error) {
 			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
+			predictOptions := []transformers.PredictOption{
+				transformers.SetTemperature(c.Temperature),
+				transformers.SetTopP(c.TopP),
+				transformers.SetTopK(c.TopK),
+				transformers.SetTokens(c.Maxtokens),
+				transformers.SetThreads(c.Threads),
 			}
 
 			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
+				predictOptions = append(predictOptions, transformers.SetBatch(c.Batch))
 			}
 
 			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
+				predictOptions = append(predictOptions, transformers.SetSeed(c.Seed))
 			}
 
 			return model.Predict(
@@ -291,23 +304,23 @@ func ModelInference(s string, loader *model.ModelLoader, c Config, tokenCallback
 				predictOptions...,
 			)
 		}
-	case *gpt2.Starcoder:
+	case *transformers.Starcoder:
 		fn = func() (string, error) {
 			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
+			predictOptions := []transformers.PredictOption{
+				transformers.SetTemperature(c.Temperature),
+				transformers.SetTopP(c.TopP),
+				transformers.SetTopK(c.TopK),
+				transformers.SetTokens(c.Maxtokens),
+				transformers.SetThreads(c.Threads),
 			}
 
 			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
+				predictOptions = append(predictOptions, transformers.SetBatch(c.Batch))
 			}
 
 			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
+				predictOptions = append(predictOptions, transformers.SetSeed(c.Seed))
 			}
 
 			return model.Predict(
@@ -315,23 +328,23 @@ func ModelInference(s string, loader *model.ModelLoader, c Config, tokenCallback
 				predictOptions...,
 			)
 		}
-	case *gpt2.RedPajama:
+	case *transformers.MPT:
 		fn = func() (string, error) {
 			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
+			predictOptions := []transformers.PredictOption{
+				transformers.SetTemperature(c.Temperature),
+				transformers.SetTopP(c.TopP),
+				transformers.SetTopK(c.TopK),
+				transformers.SetTokens(c.Maxtokens),
+				transformers.SetThreads(c.Threads),
 			}
 
 			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
+				predictOptions = append(predictOptions, transformers.SetBatch(c.Batch))
 			}
 
 			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
+				predictOptions = append(predictOptions, transformers.SetSeed(c.Seed))
 			}
 
 			return model.Predict(
@@ -359,23 +372,23 @@ func ModelInference(s string, loader *model.ModelLoader, c Config, tokenCallback
 				predictOptions...,
 			)
 		}
-	case *gpt2.StableLM:
+	case *transformers.GPTJ:
 		fn = func() (string, error) {
 			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
+			predictOptions := []transformers.PredictOption{
+				transformers.SetTemperature(c.Temperature),
+				transformers.SetTopP(c.TopP),
+				transformers.SetTopK(c.TopK),
+				transformers.SetTokens(c.Maxtokens),
+				transformers.SetThreads(c.Threads),
 			}
 
 			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
+				predictOptions = append(predictOptions, transformers.SetBatch(c.Batch))
 			}
 
 			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
+				predictOptions = append(predictOptions, transformers.SetSeed(c.Seed))
 			}
 
 			return model.Predict(
@@ -383,23 +396,23 @@ func ModelInference(s string, loader *model.ModelLoader, c Config, tokenCallback
 				predictOptions...,
 			)
 		}
-	case *gpt2.Dolly:
+	case *transformers.Dolly:
 		fn = func() (string, error) {
 			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
+			predictOptions := []transformers.PredictOption{
+				transformers.SetTemperature(c.Temperature),
+				transformers.SetTopP(c.TopP),
+				transformers.SetTopK(c.TopK),
+				transformers.SetTokens(c.Maxtokens),
+				transformers.SetThreads(c.Threads),
 			}
 
 			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
+				predictOptions = append(predictOptions, transformers.SetBatch(c.Batch))
 			}
 
 			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
+				predictOptions = append(predictOptions, transformers.SetSeed(c.Seed))
 			}
 
 			return model.Predict(
@@ -407,23 +420,23 @@ func ModelInference(s string, loader *model.ModelLoader, c Config, tokenCallback
 				predictOptions...,
 			)
 		}
-	case *gpt2.GPT2:
+	case *transformers.GPT2:
 		fn = func() (string, error) {
 			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
+			predictOptions := []transformers.PredictOption{
+				transformers.SetTemperature(c.Temperature),
+				transformers.SetTopP(c.TopP),
+				transformers.SetTopK(c.TopK),
+				transformers.SetTokens(c.Maxtokens),
+				transformers.SetThreads(c.Threads),
 			}
 
 			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
+				predictOptions = append(predictOptions, transformers.SetBatch(c.Batch))
 			}
 
 			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
+				predictOptions = append(predictOptions, transformers.SetSeed(c.Seed))
 			}
 
 			return model.Predict(
@@ -469,7 +482,7 @@ func ModelInference(s string, loader *model.ModelLoader, c Config, tokenCallback
 				model.SetTokenCallback(tokenCallback)
 			}
 
-			predictOptions := buildLLamaPredictOptions(c)
+			predictOptions := buildLLamaPredictOptions(c, loader.ModelPath)
 
 			str, er := model.Predict(
 				s,

examples/README.md

@@ -4,6 +4,13 @@ Here is a list of projects that can easily be integrated with the LocalAI backen
 
 ### Projects
 
+### AutoGPT
+
+_by [@mudler](https://github.com/mudler)_
+
+This example shows how to use AutoGPT with LocalAI.
+
+[Check it out here](https://github.com/go-skynet/LocalAI/tree/master/examples/autoGPT/)
 
 ### Chatbot-UI
 
@@ -57,6 +64,14 @@ A full example on how to run RWKV models with LocalAI
 
 [Check it out here](https://github.com/go-skynet/LocalAI/tree/master/examples/rwkv/)
 
+### PrivateGPT
+
+_by [@mudler](https://github.com/mudler)_
+
+A full example on how to run PrivateGPT with LocalAI
+
+[Check it out here](https://github.com/go-skynet/LocalAI/tree/master/examples/privateGPT/)
+
 ### Slack bot
 
 _by [@mudler](https://github.com/mudler)_

examples/autoGPT/.env

@@ -0,0 +1,5 @@
+OPENAI_API_KEY=sk---anystringhere
+OPENAI_API_BASE=http://api:8080/v1
+# Models to preload at start
+# Here we configure gpt4all as gpt-3.5-turbo and bert as embeddings
+PRELOAD_MODELS=[{"url": "github:go-skynet/model-gallery/gpt4all-j.yaml", "name": "gpt-3.5-turbo"}, { "url": "github:go-skynet/model-gallery/bert-embeddings.yaml", "name": "text-embedding-ada-002"}]

examples/autoGPT/README.md

@@ -0,0 +1,32 @@
+# AutoGPT
+
+Example of integration with [AutoGPT](https://github.com/Significant-Gravitas/Auto-GPT).
+
+## Run
+
+```bash
+# Clone LocalAI
+git clone https://github.com/go-skynet/LocalAI
+
+cd LocalAI/examples/autoGPT
+
+docker-compose run --rm auto-gpt
+```
+
+Note: The example automatically downloads the `gpt4all` model as it is under a permissive license. The GPT4All model does not seem to be enough to run AutoGPT. WizardLM-7b-uncensored seems to perform better (with `f16: true`).
+
+See the `.env` configuration file to set a different model with the [model-gallery](https://github.com/go-skynet/model-gallery) by editing `PRELOAD_MODELS`.
+
+## Without docker
+
+Run AutoGPT with `OPENAI_API_BASE` pointing to the LocalAI endpoint. If you run it locally for instance:
+
+```
+OPENAI_API_BASE=http://localhost:8080 python ...
+```
+
+Note: you need a model named `gpt-3.5-turbo` and `text-embedding-ada-002`. You can preload those in LocalAI at start by setting in the env:
+
+```
+PRELOAD_MODELS=[{"url": "github:go-skynet/model-gallery/gpt4all-j.yaml", "name": "gpt-3.5-turbo"}, { "url": "github:go-skynet/model-gallery/bert-embeddings.yaml", "name": "text-embedding-ada-002"}]
+```

examples/autoGPT/docker-compose.yaml

@@ -0,0 +1,42 @@
+version: "3.9"
+services:
+  api:
+    image: quay.io/go-skynet/local-ai:latest
+    ports:
+      - 8080:8080
+    env_file:
+      - .env
+    environment:
+      - DEBUG=true
+      - MODELS_PATH=/models
+    volumes:
+      - ./models:/models:cached
+    command: ["/usr/bin/local-ai" ]
+  auto-gpt:
+    image: significantgravitas/auto-gpt
+    depends_on:
+      api:
+        condition: service_healthy
+      redis:
+        condition: service_started
+    env_file:
+      - .env
+    environment:
+      MEMORY_BACKEND: ${MEMORY_BACKEND:-redis}
+      REDIS_HOST: ${REDIS_HOST:-redis}
+    profiles: ["exclude-from-up"]
+    volumes:
+      - ./auto_gpt_workspace:/app/autogpt/auto_gpt_workspace
+      - ./data:/app/data
+      ## allow auto-gpt to write logs to disk
+      - ./logs:/app/logs
+      ## uncomment following lines if you want to make use of these files
+      ## you must have them existing in the same folder as this docker-compose.yml
+      #- type: bind
+      #  source: ./azure.yaml
+      #  target: /app/azure.yaml
+      #- type: bind
+      #  source: ./ai_settings.yaml
+      #  target: /app/ai_settings.yaml
+  redis:
+    image: "redis/redis-stack-server:latest"

examples/privateGPT/README.md

@@ -0,0 +1,25 @@
+# privateGPT
+
+This example is a re-adaptation of https://github.com/imartinez/privateGPT to work with LocalAI and OpenAI endpoints. We have a fork with the changes required to work with privateGPT here https://github.com/go-skynet/privateGPT ( PR: https://github.com/imartinez/privateGPT/pull/408 ).
+
+Follow the instructions in https://github.com/go-skynet/privateGPT:
+
+```bash
+git clone git@github.com:go-skynet/privateGPT.git
+cd privateGPT
+pip install -r requirements.txt
+```
+
+Rename `example.env` to `.env` and edit the variables appropriately.
+
+This is an example `.env` file for LocalAI:
+
+```
+PERSIST_DIRECTORY=db
+# Set to OpenAI here
+MODEL_TYPE=OpenAI
+EMBEDDINGS_MODEL_NAME=all-MiniLM-L6-v2
+MODEL_N_CTX=1000
+# LocalAI URL
+OPENAI_API_BASE=http://localhost:8080/v1
+```

go.mod

@@ -4,18 +4,18 @@ go 1.19
 
 require (
 	github.com/donomii/go-rwkv.cpp v0.0.0-20230515123100-6fdd0c338e56
-	github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230520182345-041be06d5881
+	github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230527074028-9b926844e3ae
 	github.com/go-audio/wav v1.1.0
 	github.com/go-skynet/bloomz.cpp v0.0.0-20230510223001-e9366e82abdf
 	github.com/go-skynet/go-bert.cpp v0.0.0-20230516063724-cea1ed76a7f4
-	github.com/go-skynet/go-gpt2.cpp v0.0.0-20230512145559-7bff56f02245
-	github.com/go-skynet/go-llama.cpp v0.0.0-20230520155239-ccf23adfb278
+	github.com/go-skynet/go-ggml-transformers.cpp v0.0.0-20230525204055-4f18e5eb7508
+	github.com/go-skynet/go-llama.cpp v0.0.0-20230524233806-6e7e69a1607e
 	github.com/gofiber/fiber/v2 v2.46.0
 	github.com/google/uuid v1.3.0
 	github.com/hashicorp/go-multierror v1.1.1
-	github.com/imdario/mergo v0.3.15
+	github.com/imdario/mergo v0.3.16
 	github.com/mudler/go-stable-diffusion v0.0.0-20230516152536-c0748eca3642
-	github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230521011615-aba1147a2253
+	github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230526132403-a6f3e94458e2
 	github.com/onsi/ginkgo/v2 v2.9.5
 	github.com/onsi/gomega v1.27.7
 	github.com/otiai10/openaigo v1.1.0
@@ -41,6 +41,7 @@ require (
 	github.com/go-openapi/jsonreference v0.19.6 // indirect
 	github.com/go-openapi/spec v0.20.4 // indirect
 	github.com/go-openapi/swag v0.19.15 // indirect
+	github.com/go-skynet/go-gpt2.cpp v0.0.0-20230523153133-3eb3a32c0874 // indirect
 	github.com/go-task/slim-sprig v0.0.0-20230315185526-52ccab3ef572 // indirect
 	github.com/google/go-cmp v0.5.9 // indirect
 	github.com/google/pprof v0.0.0-20210407192527-94a9f03dee38 // indirect

go.sum

@@ -18,12 +18,14 @@ github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c
 github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/donomii/go-rwkv.cpp v0.0.0-20230515123100-6fdd0c338e56 h1:s8/MZdicstKi5fn9D9mKGIQ/q6IWCYCk/BM68i8v51w=
 github.com/donomii/go-rwkv.cpp v0.0.0-20230515123100-6fdd0c338e56/go.mod h1:gWy7FIWioqYmYxkaoFyBnaKApeZVrUkHhv9EV9pz4dM=
-github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230515153606-95b02d76b04d h1:uxKTbiRnplE2SubchneSf4NChtxLJtOy9VdHnQMT0d0=
-github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230515153606-95b02d76b04d/go.mod h1:QIjZ9OktHFG7p+/m3sMvrAJKKdWrr1fZIK0rM6HZlyo=
-github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230520170006-429b9785c080 h1:W3itqKpRX9FhheKiAxdmuOBy/mjDfMf2G1vcuFIYqZc=
-github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230520170006-429b9785c080/go.mod h1:QIjZ9OktHFG7p+/m3sMvrAJKKdWrr1fZIK0rM6HZlyo=
 github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230520182345-041be06d5881 h1:dafqVivljYk51VLFnnpTXJnfWDe637EobWZ1l8PyEf8=
 github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230520182345-041be06d5881/go.mod h1:QIjZ9OktHFG7p+/m3sMvrAJKKdWrr1fZIK0rM6HZlyo=
+github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230523110439-77eab3fbfe5e h1:4PMorQuoUGAXmIzCtnNOHaasyLokXdgd8jUWwsraFTo=
+github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230523110439-77eab3fbfe5e/go.mod h1:QIjZ9OktHFG7p+/m3sMvrAJKKdWrr1fZIK0rM6HZlyo=
+github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230524181101-5e2b3407ef46 h1:+STJWsBFikYC90LnR8I9gcBdysQn7Jv9Jb44+5WBi68=
+github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230524181101-5e2b3407ef46/go.mod h1:QIjZ9OktHFG7p+/m3sMvrAJKKdWrr1fZIK0rM6HZlyo=
+github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230527074028-9b926844e3ae h1:uzi5myq/qNX9xiKMRF/fW3HfxuEo2WcnTalwg9fe2hM=
+github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230527074028-9b926844e3ae/go.mod h1:QIjZ9OktHFG7p+/m3sMvrAJKKdWrr1fZIK0rM6HZlyo=
 github.com/go-audio/audio v1.0.0 h1:zS9vebldgbQqktK4H0lUqWrG8P0NxCJVqcj7ZpNnwd4=
 github.com/go-audio/audio v1.0.0/go.mod h1:6uAu0+H2lHkwdGsAY+j2wHPNPpPoeg5AaEFh9FlA+Zs=
 github.com/go-audio/riff v1.0.0 h1:d8iCGbDvox9BfLagY94fBynxSPHO80LmZCaOsmKxokA=
@@ -46,17 +48,23 @@ github.com/go-skynet/bloomz.cpp v0.0.0-20230510223001-e9366e82abdf h1:VJfSn8hIDE
 github.com/go-skynet/bloomz.cpp v0.0.0-20230510223001-e9366e82abdf/go.mod h1:wc0fJ9V04yiYTfgKvE5RUUSRQ5Kzi0Bo4I+U3nNOUuA=
 github.com/go-skynet/go-bert.cpp v0.0.0-20230516063724-cea1ed76a7f4 h1:+3KPDf4Wv1VHOkzAfZnlj9qakLSYggTpm80AswhD/FU=
 github.com/go-skynet/go-bert.cpp v0.0.0-20230516063724-cea1ed76a7f4/go.mod h1:VY0s5KoAI2jRCvQXKuDeEEe8KG7VaWifSNJSk+E1KtY=
-github.com/go-skynet/go-gpt2.cpp v0.0.0-20230512145559-7bff56f02245 h1:IcfYY5uH0DdDXEJKJ8bq0WZCd9guPPd3xllaWNy8LOk=
-github.com/go-skynet/go-gpt2.cpp v0.0.0-20230512145559-7bff56f02245/go.mod h1:1Wj/xbkMfwQSOrhNYK178IzqQHstZbRfhx4s8p1M5VM=
-github.com/go-skynet/go-llama.cpp v0.0.0-20230520082618-a298043ef5f1 h1:i0oM2MERUgMIRmjOcv22TDQULxbmY8o9rZKLKKyWXLo=
-github.com/go-skynet/go-llama.cpp v0.0.0-20230520082618-a298043ef5f1/go.mod h1:oA0r4BW8ndyjTMGi1tulsNd7sdg3Ql8MaVFuT1zF6ws=
+github.com/go-skynet/go-ggml-transformers.cpp v0.0.0-20230523173010-f89d7c22df6b h1:uKICsAbdRJxMPZ4RXltwOwXPRDO1/d/pdGR3gEEUV9M=
+github.com/go-skynet/go-ggml-transformers.cpp v0.0.0-20230523173010-f89d7c22df6b/go.mod h1:hjmO5UfipWl6xkPT54acOs9DDto8GPV81IvsBcvRjsA=
+github.com/go-skynet/go-ggml-transformers.cpp v0.0.0-20230524084634-c4c581f1853c h1:jXUOCh2K4OzRItTtHzdxvkylE9r1szRSleRpXCNvraY=
+github.com/go-skynet/go-ggml-transformers.cpp v0.0.0-20230524084634-c4c581f1853c/go.mod h1:hjmO5UfipWl6xkPT54acOs9DDto8GPV81IvsBcvRjsA=
+github.com/go-skynet/go-ggml-transformers.cpp v0.0.0-20230525204055-4f18e5eb7508 h1:pb7wUQlgqbakB4vILBq44iLe5w9bcjAsP7js2iFOWX8=
+github.com/go-skynet/go-ggml-transformers.cpp v0.0.0-20230525204055-4f18e5eb7508/go.mod h1:hjmO5UfipWl6xkPT54acOs9DDto8GPV81IvsBcvRjsA=
+github.com/go-skynet/go-gpt2.cpp v0.0.0-20230523153133-3eb3a32c0874 h1:/6QWh2oarU7iPSpXj/3bLlkKptyxjKTRrNtGUrh8vhI=
+github.com/go-skynet/go-gpt2.cpp v0.0.0-20230523153133-3eb3a32c0874/go.mod h1:1Wj/xbkMfwQSOrhNYK178IzqQHstZbRfhx4s8p1M5VM=
 github.com/go-skynet/go-llama.cpp v0.0.0-20230520155239-ccf23adfb278 h1:st4ow9JKy3UuhkwutrbWof2vMFU/YxwBCLYZ1IxJ2Po=
 github.com/go-skynet/go-llama.cpp v0.0.0-20230520155239-ccf23adfb278/go.mod h1:oA0r4BW8ndyjTMGi1tulsNd7sdg3Ql8MaVFuT1zF6ws=
+github.com/go-skynet/go-llama.cpp v0.0.0-20230523103108-dcf8da632bce h1:Mcq9LvYG4msXJvFUeiYI6PGftqmYbOoBxNfjyAAyFB4=
+github.com/go-skynet/go-llama.cpp v0.0.0-20230523103108-dcf8da632bce/go.mod h1:oA0r4BW8ndyjTMGi1tulsNd7sdg3Ql8MaVFuT1zF6ws=
+github.com/go-skynet/go-llama.cpp v0.0.0-20230524233806-6e7e69a1607e h1:zfxPbHj7/hN2F7V12vfxCi4CFsaVO1WohW96OVFtfNw=
+github.com/go-skynet/go-llama.cpp v0.0.0-20230524233806-6e7e69a1607e/go.mod h1:oA0r4BW8ndyjTMGi1tulsNd7sdg3Ql8MaVFuT1zF6ws=
 github.com/go-task/slim-sprig v0.0.0-20230315185526-52ccab3ef572 h1:tfuBGBXKqDEevZMzYi5KSi8KkcZtzBcTgAUUtapy0OI=
 github.com/go-task/slim-sprig v0.0.0-20230315185526-52ccab3ef572/go.mod h1:9Pwr4B2jHnOSGXyyzV8ROjYa2ojvAY6HCGYYfMoC3Ls=
 github.com/godbus/dbus/v5 v5.0.4/go.mod h1:xhWf0FNVPg57R7Z0UbKHbJfkEywrmjJnf7w5xrFpKfA=
-github.com/gofiber/fiber/v2 v2.45.0 h1:p4RpkJT9GAW6parBSbcNFH2ApnAuW3OzaQzbOCoDu+s=
-github.com/gofiber/fiber/v2 v2.45.0/go.mod h1:DNl0/c37WLe0g92U6lx1VMQuxGUQY5V7EIaVoEsUffc=
 github.com/gofiber/fiber/v2 v2.46.0 h1:wkkWotblsGVlLjXj2dpgKQAYHtXumsK/HyFugQM68Ns=
 github.com/gofiber/fiber/v2 v2.46.0/go.mod h1:DNl0/c37WLe0g92U6lx1VMQuxGUQY5V7EIaVoEsUffc=
 github.com/golang/protobuf v1.5.3 h1:KhyjKVUg7Usr/dYsdSqoFveMYd5ko72D+zANwlG1mmg=
@@ -73,6 +81,8 @@ github.com/hashicorp/go-multierror v1.1.1/go.mod h1:iw975J/qwKPdAO1clOe2L8331t/9
 github.com/ianlancetaylor/demangle v0.0.0-20200824232613-28f6c0f3b639/go.mod h1:aSSvb/t6k1mPoxDqO4vJh6VOCGPwU4O0C2/Eqndh1Sc=
 github.com/imdario/mergo v0.3.15 h1:M8XP7IuFNsqUx6VPK2P9OSmsYsI/YFaGil0uD21V3dM=
 github.com/imdario/mergo v0.3.15/go.mod h1:WBLT9ZmE3lPoWsEzCh9LPo3TiwVN+ZKEjmz+hD27ysY=
+github.com/imdario/mergo v0.3.16 h1:wwQJbIsHYGMUyLSPrEq1CT16AhnhNJQ51+4fdHUnCl4=
+github.com/imdario/mergo v0.3.16/go.mod h1:WBLT9ZmE3lPoWsEzCh9LPo3TiwVN+ZKEjmz+hD27ysY=
 github.com/josharian/intern v1.0.0 h1:vlS4z54oSdjm0bgjRigI+G1HpF+tI+9rE5LLzOg8HmY=
 github.com/josharian/intern v1.0.0/go.mod h1:5DoeVV0s6jJacbCEi61lwdGj/aVlrQvzHFFd8Hwg//Y=
 github.com/klauspost/compress v1.16.3 h1:XuJt9zzcnaz6a16/OU53ZjWp/v7/42WcR5t2a0PcNQY=
@@ -99,10 +109,18 @@ github.com/mudler/go-stable-diffusion v0.0.0-20230516152536-c0748eca3642 h1:KTkh
 github.com/mudler/go-stable-diffusion v0.0.0-20230516152536-c0748eca3642/go.mod h1:8ufRkpz/S/9ahkaxzZ5i4WMgO9w4InEhuRoT7vK5Rnw=
 github.com/niemeyer/pretty v0.0.0-20200227124842-a10e7caefd8e h1:fD57ERR4JtEqsWbfPhv4DMiApHyliiK5xCTNVSPiaAs=
 github.com/niemeyer/pretty v0.0.0-20200227124842-a10e7caefd8e/go.mod h1:zD1mROLANZcx1PVRCS0qkT7pwLkGfwJo4zjcN/Tysno=
-github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230519014017-914519e772fd h1:kMnZASxCNc8GsPuAV94tltEsfT6T+esuB+rgzdjwFVM=
-github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230519014017-914519e772fd/go.mod h1:4T3CHXyrt+7FQHXaxULZfPjHbD8/99WuDDJa0YVZARI=
-github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230521011615-aba1147a2253 h1:7udNpoHYOBktcpCEe8aDaPJ0LyzyRhVjpzAGFjPxPkY=
-github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230521011615-aba1147a2253/go.mod h1:4T3CHXyrt+7FQHXaxULZfPjHbD8/99WuDDJa0YVZARI=
+github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230522220313-2ce22208a3dd h1:is/rE0YD8oEWcX3fQ+VxoS3fD0LqFEmTxh8XZegYYsA=
+github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230522220313-2ce22208a3dd/go.mod h1:4T3CHXyrt+7FQHXaxULZfPjHbD8/99WuDDJa0YVZARI=
+github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230523222017-b36a52020702 h1:uya1G35AbUfVtG8fu/HuUGTFXpN7n9XuRAAvC1lTr+M=
+github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230523222017-b36a52020702/go.mod h1:4T3CHXyrt+7FQHXaxULZfPjHbD8/99WuDDJa0YVZARI=
+github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230525153421-63f57635d83c h1:mDy1OKHlG9xv1KDMcOVNYQwoYKZSlb5Mu69W3+DNLYI=
+github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230525153421-63f57635d83c/go.mod h1:4T3CHXyrt+7FQHXaxULZfPjHbD8/99WuDDJa0YVZARI=
+github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230525202709-afe3870b7a29 h1:hgml/PMZX3M+WigXD4BGy+mbD1oPxYbXJXo16I555Aw=
+github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230525202709-afe3870b7a29/go.mod h1:4T3CHXyrt+7FQHXaxULZfPjHbD8/99WuDDJa0YVZARI=
+github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230525210850-d1ff7132c553 h1:+zQQHEoOaVUT72uLr6OJF+Lj35LR620aeeyrF7K6x5s=
+github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230525210850-d1ff7132c553/go.mod h1:4T3CHXyrt+7FQHXaxULZfPjHbD8/99WuDDJa0YVZARI=
+github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230526132403-a6f3e94458e2 h1:DE++nIPuUGk8pz71PF0BITX+CTF0lv4ZNWv12qCBUVk=
+github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230526132403-a6f3e94458e2/go.mod h1:4T3CHXyrt+7FQHXaxULZfPjHbD8/99WuDDJa0YVZARI=
 github.com/onsi/ginkgo/v2 v2.9.5 h1:+6Hr4uxzP4XIUyAkg61dWBw8lb/gc4/X5luuxN/EC+Q=
 github.com/onsi/ginkgo/v2 v2.9.5/go.mod h1:tvAoo1QUJwNEU2ITftXTpR7R1RbCzoZUOs3RonqW57k=
 github.com/onsi/gomega v1.27.7 h1:fVih9JD6ogIiHUN6ePK7HJidyEDpWGVB5mzM7cWNXoU=

main.go

@@ -53,6 +53,16 @@ func main() {
 				EnvVars:     []string{"MODELS_PATH"},
 				Value:       filepath.Join(path, "models"),
 			},
+			&cli.StringFlag{
+				Name:        "preload-models",
+				DefaultText: "A List of models to apply in JSON at start",
+				EnvVars:     []string{"PRELOAD_MODELS"},
+			},
+			&cli.StringFlag{
+				Name:        "preload-models-config",
+				DefaultText: "A List of models to apply at startup. Path to a YAML config file",
+				EnvVars:     []string{"PRELOAD_MODELS_CONFIG"},
+			},
 			&cli.StringFlag{
 				Name:        "config-file",
 				DefaultText: "Config file",
@@ -103,11 +113,14 @@ It uses llama.cpp, ggml and gpt4all as backend with golang c bindings.
 			fmt.Printf("Starting LocalAI using %d threads, with models path: %s\n", ctx.Int("threads"), ctx.String("models-path"))
 			return api.App(
 				api.WithConfigFile(ctx.String("config-file")),
+				api.WithJSONStringPreload(ctx.String("preload-models")),
+				api.WithYAMLConfigPreload(ctx.String("preload-models-config")),
 				api.WithModelLoader(model.NewModelLoader(ctx.String("models-path"))),
 				api.WithContextSize(ctx.Int("context-size")),
 				api.WithDebug(ctx.Bool("debug")),
 				api.WithImageDir(ctx.String("image-path")),
 				api.WithF16(ctx.Bool("f16")),
+				api.WithDisableMessage(false),
 				api.WithCors(ctx.Bool("cors")),
 				api.WithCorsAllowOrigins(ctx.String("cors-allow-origins")),
 				api.WithThreads(ctx.Int("threads")),

pkg/model/initializers.go

@@ -10,7 +10,7 @@ import (
 	"github.com/go-skynet/LocalAI/pkg/stablediffusion"
 	bloomz "github.com/go-skynet/bloomz.cpp"
 	bert "github.com/go-skynet/go-bert.cpp"
-	gpt2 "github.com/go-skynet/go-gpt2.cpp"
+	transformers "github.com/go-skynet/go-ggml-transformers.cpp"
 	llama "github.com/go-skynet/go-llama.cpp"
 	"github.com/hashicorp/go-multierror"
 	gpt4all "github.com/nomic-ai/gpt4all/gpt4all-bindings/golang"
@@ -23,9 +23,9 @@ const (
 	LlamaBackend           = "llama"
 	BloomzBackend          = "bloomz"
 	StarcoderBackend       = "starcoder"
-	StableLMBackend        = "stablelm"
+	GPTJBackend            = "gptj"
 	DollyBackend           = "dolly"
-	RedPajamaBackend       = "redpajama"
+	MPTBackend             = "mpt"
 	GPTNeoXBackend         = "gptneox"
 	ReplitBackend          = "replit"
 	Gpt2Backend            = "gpt2"
@@ -43,41 +43,41 @@ var backends []string = []string{
 	Gpt4AllLlamaBackend,
 	Gpt4AllMptBackend,
 	Gpt4AllJBackend,
-	Gpt2Backend,
-	WhisperBackend,
 	RwkvBackend,
-	BloomzBackend,
-	StableLMBackend,
-	DollyBackend,
-	RedPajamaBackend,
-	ReplitBackend,
 	GPTNeoXBackend,
+	WhisperBackend,
 	BertEmbeddingsBackend,
+	GPTJBackend,
+	Gpt2Backend,
+	DollyBackend,
+	MPTBackend,
+	ReplitBackend,
 	StarcoderBackend,
+	BloomzBackend,
 }
 
 var starCoder = func(modelFile string) (interface{}, error) {
-	return gpt2.NewStarcoder(modelFile)
+	return transformers.NewStarcoder(modelFile)
 }
 
-var redPajama = func(modelFile string) (interface{}, error) {
-	return gpt2.NewRedPajama(modelFile)
+var mpt = func(modelFile string) (interface{}, error) {
+	return transformers.NewMPT(modelFile)
 }
 
 var dolly = func(modelFile string) (interface{}, error) {
-	return gpt2.NewDolly(modelFile)
+	return transformers.NewDolly(modelFile)
 }
 
 var gptNeoX = func(modelFile string) (interface{}, error) {
-	return gpt2.NewGPTNeoX(modelFile)
+	return transformers.NewGPTNeoX(modelFile)
 }
 
 var replit = func(modelFile string) (interface{}, error) {
-	return gpt2.NewReplit(modelFile)
+	return transformers.NewReplit(modelFile)
 }
 
-var stableLM = func(modelFile string) (interface{}, error) {
-	return gpt2.NewStableLM(modelFile)
+var gptJ = func(modelFile string) (interface{}, error) {
+	return transformers.NewGPTJ(modelFile)
 }
 
 var bertEmbeddings = func(modelFile string) (interface{}, error) {
@@ -87,8 +87,9 @@ var bertEmbeddings = func(modelFile string) (interface{}, error) {
 var bloomzLM = func(modelFile string) (interface{}, error) {
 	return bloomz.New(modelFile)
 }
-var gpt2LM = func(modelFile string) (interface{}, error) {
-	return gpt2.New(modelFile)
+
+var transformersLM = func(modelFile string) (interface{}, error) {
+	return transformers.New(modelFile)
 }
 
 var stableDiffusion = func(assetDir string) (interface{}, error) {
@@ -130,14 +131,14 @@ func (ml *ModelLoader) BackendLoader(backendString string, modelFile string, lla
 		return ml.LoadModel(modelFile, llamaLM(llamaOpts...))
 	case BloomzBackend:
 		return ml.LoadModel(modelFile, bloomzLM)
-	case StableLMBackend:
-		return ml.LoadModel(modelFile, stableLM)
+	case GPTJBackend:
+		return ml.LoadModel(modelFile, gptJ)
 	case DollyBackend:
 		return ml.LoadModel(modelFile, dolly)
-	case RedPajamaBackend:
-		return ml.LoadModel(modelFile, redPajama)
+	case MPTBackend:
+		return ml.LoadModel(modelFile, mpt)
 	case Gpt2Backend:
-		return ml.LoadModel(modelFile, gpt2LM)
+		return ml.LoadModel(modelFile, transformersLM)
 	case GPTNeoXBackend:
 		return ml.LoadModel(modelFile, gptNeoX)
 	case ReplitBackend:

tests/models_fixtures/embeddings.yaml

@@ -1,6 +1,5 @@
 name: text-embedding-ada-002
 parameters:
   model: bert
-threads: 14
 backend: bert-embeddings
 embeddings: true

tests/models_fixtures/rwkv.yaml

@@ -6,7 +6,6 @@ parameters:
   max_tokens: 100
   top_p: 0.8
 context_size: 1024
-threads: 14
 backend: "rwkv"
 cutwords:
 - "Bob:.*"