From d1ceeaa99a2791efc4bcc533c23b3e3fc12ed6bb Mon Sep 17 00:00:00 2001 From: mudler-agent Date: Fri, 18 Sep 2026 23:24:36 +0200 Subject: [PATCH] feat(gallery): consolidate 25 pending gallery PRs (#12124) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Consolidates all 25 pending gallery bot PRs into a single merge to resolve the conflict cascade — every PR branched from a different point in master and they all touch gallery/index.yaml, so merging them individually was blocked by constant conflicts. Changes: - gallery/index.yaml: +739 lines (new model entries and fixes) - docs/content/features/model-gallery.md: +116 lines (new model docs) - docs/content/features/audio-cpp.md: +12 lines (Sortformer checksum fix) Entry count: 1597 -> 1890 (293 new entries, no duplicates, YAML validated). Supersedes: #11986 #11992 #11994 #11996 #11999 #12002 #12017 #12019 #12021 #12025 #12027 #12029 #12032 #12036 #12037 #12038 #12041 #12042 #12043 #12047 #12050 #12064 #12065 #12066 #12118 Assisted-by: MAKI:regolo/glm5.2 Signed-off-by: Ettore Di Giacinto Co-authored-by: Ettore Di Giacinto --- docs/content/features/audio-cpp.md | 12 + docs/content/features/model-gallery.md | 116 ++++ gallery/index.yaml | 739 ++++++++++++++++++++++--- 3 files changed, 780 insertions(+), 87 deletions(-) diff --git a/docs/content/features/audio-cpp.md b/docs/content/features/audio-cpp.md index f3a7ce4be..2adfa47d7 100644 --- a/docs/content/features/audio-cpp.md +++ b/docs/content/features/audio-cpp.md @@ -33,6 +33,18 @@ an available compute backend, with CPU as the fallback. To force CPU execution, replace that option with `backend:cpu`. Model configurations that omit this option still default to CPU. +### Sortformer installation + +Install the Sortformer model for speaker diarization from the gallery: + +```bash +local-ai models install audio-cpp-sortformer-diarization +``` + +If installation fails with `SHA mismatch`, compare the reported metadata checksum +with the current gallery entry. An outdated gallery checksum can reject a complete +download. Report a persistent mismatch with both checksum values and the model name. + ## What it serves One model serves one family, and a family advertises the tasks it can perform. The diff --git a/docs/content/features/model-gallery.md b/docs/content/features/model-gallery.md index f0088f583..bc8e524c6 100644 --- a/docs/content/features/model-gallery.md +++ b/docs/content/features/model-gallery.md @@ -23,6 +23,17 @@ GPT and text generation models might have a license which is not permissive for ![output](https://github.com/mudler/LocalAI/assets/2420543/7b16676e-d5b1-4c97-89bd-9fa5065c21ad) +## Instella-MoE availability + +The gallery excludes `instella-moe-16b-a3b-think` and +`instella-moe-16b-a3b-think-q8` because the packaged llama.cpp backend does not +support their `instella-moe` architecture. Loading these GGUF files fails with +`unknown model architecture: 'instella-moe'`. + +The entries can return after LocalAI ships a compatible backend. See +[the compatibility issue](https://github.com/mudler/LocalAI/issues/11681) and +[upstream llama.cpp support](https://github.com/ggml-org/llama.cpp/pull/26467). + ## Useful Links and resources - [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) - here you can find a list of the most performing models on the Open LLM benchmark. Keep in mind models compatible with LocalAI must be quantized in the `gguf` format. @@ -43,6 +54,25 @@ Both views use the same model selection and store the view, search, filter, and selection in the URL. Installing from Explore does not move you away from the catalog; the entry updates in place when the operation finishes. +## NeoHorse-1-4B + +Install NeoHorse-1-4B with automatic selection between Q4_K_M, Q6_K, and Q8_0 GGUF builds: + +```bash +local-ai models install neohorse-1-4b-q4 +``` + +To select Q8_0 explicitly: + +```bash +local-ai models install neohorse-1-4b-q4 --variant neohorse-1-4b-q8 +``` + +[NeoHorse-1-4B](https://huggingface.co/TokenRhythm/NeoHorse-1-4B) is a text-only Qwen3.5 fine-tune for coding, reasoning, and agentic tasks. +These builds use llama.cpp and the embedded Jinja chat template. +The gallery defaults to 32,768 context tokens; the model supports up to 262,144 tokens with sufficient memory. +The [GGUF downloads](https://huggingface.co/mradermacher/NeoHorse-1-4B-GGUF) are pinned to a revision and verified with SHA256 checksums. + ## Spark-X2.5-1.7B Install Spark-X2.5-1.7B with automatic selection between its Q4_K_M and Q8_0 @@ -64,6 +94,19 @@ The [source model](https://huggingface.co/XHToken/Spark-X2.5-1.7B) supports up t 1,048,576 tokens; larger contexts require more memory. Use a current LocalAI llama.cpp backend with Spark-X2.5 support. +## MiniCPM5-2B + +MiniCPM5-2B offers the official Q4_K_M, Q8_0, and F16 GGUF builds through +llama.cpp. To install the F16 build explicitly: + +```bash +local-ai models install minicpm5-2b --variant minicpm5-2b-f16 +``` + +The F16 weights require a 5.04 GB download, plus additional memory for inference. +This entry uses the embedded chat template and an 8,192-token context. +See the [official GGUF repository](https://huggingface.co/openbmb/MiniCPM5-2B-GGUF). + ## VRAM and download size estimates When browsing the gallery or importing a model by URI, LocalAI can show **estimated download size** and **estimated VRAM** for models. @@ -73,6 +116,26 @@ When browsing the gallery or importing a model by URI, LocalAI can show **estima - **Hardware fit indicator**: When your system reports GPU or RAM capacity, the gallery shows whether the estimated VRAM fits (green) or may not fit (red) using a 95% headroom rule. - Estimates are best-effort and may be missing if the server does not support HEAD/Range or the request times out. +## Gemma 4 Ortenzya 31B + +[Ortenzya](https://huggingface.co/llmfan46/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-GGUF) +is a Gemma 4 fine-tune for creative writing, translation, and roleplay. +Install it with automatic selection between Q4_K_M and Q8_0: + +```bash +local-ai models install gemma-4-31b-ortenzya-q4 +``` + +To select Q8_0 explicitly: + +```bash +local-ai models install gemma-4-31b-ortenzya-q4 --variant gemma-4-31b-ortenzya-q8 +``` + +Both builds use llama.cpp, the embedded Jinja chat template, and the BF16 +vision projector for image prompts. The gallery uses a 32,768-token context +to limit memory use, with temperature 1, top_p 0.95, and top_k 64. + ## Gemma 4 12B IT Install `gemma-4-12b-it-q4` for chat, tool use, and image prompts with @@ -90,6 +153,28 @@ temperature 1, top_k 64, and top_p 0.95. They are separate from the existing QAT builds. See the [source model](https://huggingface.co/google/gemma-4-12B-it) and [GGUF files](https://huggingface.co/unsloth/gemma-4-12b-it-GGUF). +## Ornith 1.5 9B Uncensored + +Junafinity's Ornith 1.5 9B Uncensored offers Q4_K_M, Q6_K, and Q8_0 GGUF +builds for llama.cpp. Each includes the F16 vision projector for image input. +Install with automatic variant selection: + +```bash +local-ai models install ornith-1.5-9b-uncensored-q4 +``` + +Select Q8 explicitly: + +```bash +local-ai models install ornith-1.5-9b-uncensored-q4 --variant ornith-1.5-9b-uncensored-q8 +``` + +These entries use the embedded chat template and a 32,768-token context. +Sampling defaults are temperature 0.6, top_p 0.95, top_k 20, min_p 0, +and repeat_penalty 1. The model has no MTP heads. +See the [source model](https://huggingface.co/junafinity/Ornith-1.5-9B-uncensored) +and [GGUF files](https://huggingface.co/mradermacher/Ornith-1.5-9B-uncensored-GGUF). + ## Add other galleries You can add other galleries by: @@ -412,6 +497,13 @@ The same option exists on the CLI: local-ai models install nanbeige4.1-3b-q4 --variant nanbeige4.1-3b-q8 ``` +Ornith-1.5-35B-A3B offers Q4_K_M, Q5_K_M, Q6_K, and Q8_0 GGUF builds +with a BF16 vision projector. For example, select Q6_K with: + +```bash +local-ai models install ornith-1.5-35b-a3b-q4 --variant ornith-1.5-35b-a3b-q6 +``` + The `install_model` MCP tool takes the same `variant` argument, so an assistant managing installs conversationally can pick a build too. @@ -821,3 +913,27 @@ is still running is reported as queued until the installer picks it up: A job ID is queryable from the moment `/models/apply` returns it, so a `404`/`500` from this endpoint means the ID is genuinely unknown rather than merely waiting its turn. + +### Genesis Hermes Final + +Install the Qwen3.6-35B-A3B Genesis Hermes Final model for text chat, +function calling, and image input: + +```bash +local-ai models install qwen3.6-35b-a3b-genesis-hermes-final +``` + +The gallery offers APEX, APEX Compact, and Q8_K_P builds for llama.cpp. +APEX and APEX Compact also have variants with multi-token prediction (MTP) +enabled. Each build includes the F16 vision projector and uses the embedded +Jinja chat template. LocalAI selects a variant according to available memory +and serving features. To select the plain APEX build explicitly: + +```bash +local-ai models install --variant qwen3.6-35b-a3b-genesis-hermes-final qwen3.6-35b-a3b-genesis-hermes-final +``` + +These entries set a 131,072-token context, following the +[model card's guidance](https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-Hermes-Final-GGUF) +for thinking mode. This context requires additional memory beyond the weights. +The uncensored model uses the Apache-2.0 license. diff --git a/gallery/index.yaml b/gallery/index.yaml index 857019c46..8466aa6b2 100644 --- a/gallery/index.yaml +++ b/gallery/index.yaml @@ -3358,6 +3358,135 @@ - filename: llama-cpp/mmproj/ornith-1.5-9b/mmproj-BF16.gguf uri: huggingface://ornith-ai/Ornith-1.5-9B-GGUF/mmproj-Ornith-1.5-9B-BF16.gguf sha256: 626f9f90627402a6bf4a999111d0fbd69b5fcca7aa8ba089d69e5f10e8858e1d +- &ornith-1-5-9b-uncensored + name: ornith-1.5-9b-uncensored-q4 + url: github:mudler/LocalAI/gallery/virtual.yaml@master + urls: + - https://huggingface.co/junafinity/Ornith-1.5-9B-uncensored + - https://huggingface.co/mradermacher/Ornith-1.5-9B-uncensored-GGUF + license: apache-2.0 + tags: + - llm + - gguf + - cpu + - gpu + - qwen + - reasoning + - vision + variants: + - model: ornith-1.5-9b-uncensored-q6 + - model: ornith-1.5-9b-uncensored-q8 + description: | + Junafinity's Ornith-1.5-9B Uncensored is a 9B-parameter Qwen3.5 + derivative with refusal-direction ablation for research and evaluation. + This Q4_K_M GGUF build includes the F16 vision projector and uses + the embedded chat template for text and image conversations. + overrides: + backend: llama-cpp + context_size: 32768 + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - vision + mmproj: llama-cpp/mmproj/ornith-1.5-9b-uncensored/Ornith-1.5-9B-uncensored.mmproj-f16.gguf + options: + - use_jinja:true + parameters: + model: llama-cpp/models/ornith-1.5-9b-uncensored/Ornith-1.5-9B-uncensored.Q4_K_M.gguf + temperature: 0.6 + top_p: 0.95 + top_k: 20 + min_p: 0 + repeat_penalty: 1 + template: + use_tokenizer_template: true + files: + - filename: llama-cpp/models/ornith-1.5-9b-uncensored/Ornith-1.5-9B-uncensored.Q4_K_M.gguf + uri: https://huggingface.co/mradermacher/Ornith-1.5-9B-uncensored-GGUF/resolve/1041d1eabbce89549c1b6e28b52e972093e832a1/Ornith-1.5-9B-uncensored.Q4_K_M.gguf + sha256: ee180b66bc2dcd49f41e0d3c991d2eaab3078d651d97595852d16ae6d9af3c9e + - filename: llama-cpp/mmproj/ornith-1.5-9b-uncensored/Ornith-1.5-9B-uncensored.mmproj-f16.gguf + uri: https://huggingface.co/mradermacher/Ornith-1.5-9B-uncensored-GGUF/resolve/1041d1eabbce89549c1b6e28b52e972093e832a1/Ornith-1.5-9B-uncensored.mmproj-f16.gguf + sha256: 3170dd53d7ed442c944cfb378535e96a5981283d65ec4f6c389f07999f32b380 + +- !!merge <<: *ornith-1-5-9b-uncensored + name: ornith-1.5-9b-uncensored-q6 + variants: [] + description: | + Junafinity's Ornith-1.5-9B Uncensored is a 9B-parameter Qwen3.5 + derivative with refusal-direction ablation for research and evaluation. + This Q6_K GGUF build includes the F16 vision projector and uses + the embedded chat template for text and image conversations. + overrides: + backend: llama-cpp + context_size: 32768 + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - vision + mmproj: llama-cpp/mmproj/ornith-1.5-9b-uncensored/Ornith-1.5-9B-uncensored.mmproj-f16.gguf + options: + - use_jinja:true + parameters: + model: llama-cpp/models/ornith-1.5-9b-uncensored/Ornith-1.5-9B-uncensored.Q6_K.gguf + temperature: 0.6 + top_p: 0.95 + top_k: 20 + min_p: 0 + repeat_penalty: 1 + template: + use_tokenizer_template: true + files: + - filename: llama-cpp/models/ornith-1.5-9b-uncensored/Ornith-1.5-9B-uncensored.Q6_K.gguf + uri: https://huggingface.co/mradermacher/Ornith-1.5-9B-uncensored-GGUF/resolve/1041d1eabbce89549c1b6e28b52e972093e832a1/Ornith-1.5-9B-uncensored.Q6_K.gguf + sha256: 5d48d49aae8aa7cb415e2cc4989457a094ff5a13bbb425222f1450abcd6cf1b8 + - filename: llama-cpp/mmproj/ornith-1.5-9b-uncensored/Ornith-1.5-9B-uncensored.mmproj-f16.gguf + uri: https://huggingface.co/mradermacher/Ornith-1.5-9B-uncensored-GGUF/resolve/1041d1eabbce89549c1b6e28b52e972093e832a1/Ornith-1.5-9B-uncensored.mmproj-f16.gguf + sha256: 3170dd53d7ed442c944cfb378535e96a5981283d65ec4f6c389f07999f32b380 + +- !!merge <<: *ornith-1-5-9b-uncensored + name: ornith-1.5-9b-uncensored-q8 + variants: [] + description: | + Junafinity's Ornith-1.5-9B Uncensored is a 9B-parameter Qwen3.5 + derivative with refusal-direction ablation for research and evaluation. + This Q8_0 GGUF build includes the F16 vision projector and uses + the embedded chat template for text and image conversations. + overrides: + backend: llama-cpp + context_size: 32768 + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - vision + mmproj: llama-cpp/mmproj/ornith-1.5-9b-uncensored/Ornith-1.5-9B-uncensored.mmproj-f16.gguf + options: + - use_jinja:true + parameters: + model: llama-cpp/models/ornith-1.5-9b-uncensored/Ornith-1.5-9B-uncensored.Q8_0.gguf + temperature: 0.6 + top_p: 0.95 + top_k: 20 + min_p: 0 + repeat_penalty: 1 + template: + use_tokenizer_template: true + files: + - filename: llama-cpp/models/ornith-1.5-9b-uncensored/Ornith-1.5-9B-uncensored.Q8_0.gguf + uri: https://huggingface.co/mradermacher/Ornith-1.5-9B-uncensored-GGUF/resolve/1041d1eabbce89549c1b6e28b52e972093e832a1/Ornith-1.5-9B-uncensored.Q8_0.gguf + sha256: 05e0878744b36b7c55385ecb8aaf47bddf0b43f720dbfac5c79e489c02abca8f + - filename: llama-cpp/mmproj/ornith-1.5-9b-uncensored/Ornith-1.5-9B-uncensored.mmproj-f16.gguf + uri: https://huggingface.co/mradermacher/Ornith-1.5-9B-uncensored-GGUF/resolve/1041d1eabbce89549c1b6e28b52e972093e832a1/Ornith-1.5-9B-uncensored.mmproj-f16.gguf + sha256: 3170dd53d7ed442c944cfb378535e96a5981283d65ec4f6c389f07999f32b380 + - &ornith-1-5-9b-obliterated name: "ornith-1.5-9b-obliterated-q4" variants: @@ -3450,6 +3579,8 @@ - &ornith-1-5-35b-a3b name: "ornith-1.5-35b-a3b-q4" variants: + - model: ornith-1.5-35b-a3b-q5 + - model: ornith-1.5-35b-a3b-q6 - model: ornith-1.5-35b-a3b-q8 url: "github:mudler/LocalAI/gallery/virtual.yaml@master" urls: @@ -3461,8 +3592,8 @@ tasks, and tool use. It activates about 3B parameters per token and supports text and image input with a context window of 262K tokens. - This default entry uses the Q4_K_M GGUF and BF16 vision projector. A - higher-quality Q8_0 model is available as a variant. + This default entry uses the Q4_K_M GGUF and BF16 vision projector. + Q5_K_M, Q6_K, and Q8_0 builds are available as variants. license: "mit" tags: - llm @@ -3544,6 +3675,78 @@ - filename: llama-cpp/mmproj/ornith-1.5-35b-a3b/mmproj-BF16.gguf uri: huggingface://ornith-ai/Ornith-1.5-35B-A3B-GGUF/mmproj-Ornith-1.5-35B-BF16.gguf sha256: 1921a36a85aee56cd2abd27f46701802c9d85a33474792e600df6c3b282a135d +- !!merge <<: *ornith-1-5-35b-a3b + name: "ornith-1.5-35b-a3b-q5" + variants: [] + last_checked: "2026-09-14" + description: | + Ornith-1.5-35B-A3B in the Q5_K_M GGUF format, with the shared + BF16 vision projector for multimodal prompts. + overrides: + backend: llama-cpp + context_size: 262144 + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - vision + mmproj: llama-cpp/mmproj/ornith-1.5-35b-a3b/mmproj-BF16.gguf + options: + - use_jinja:true + parameters: + min_p: 0 + model: llama-cpp/models/ornith-1.5-35b-a3b/Ornith-1.5-35B-Q5_K_M.gguf + repeat_penalty: 1 + temperature: 0.6 + top_k: 20 + top_p: 0.95 + template: + use_tokenizer_template: true + files: + - filename: llama-cpp/models/ornith-1.5-35b-a3b/Ornith-1.5-35B-Q5_K_M.gguf + uri: https://huggingface.co/ornith-ai/Ornith-1.5-35B-A3B-GGUF/resolve/12393612fd4f730ff5aadc23e9b8f9648aa49ceb/Ornith-1.5-35B-Q5_K_M.gguf + sha256: 91df97de5845100e850b4b5ec5ff35695382020b880fad6f7f51787b3a953bd0 + - filename: llama-cpp/mmproj/ornith-1.5-35b-a3b/mmproj-BF16.gguf + uri: https://huggingface.co/ornith-ai/Ornith-1.5-35B-A3B-GGUF/resolve/12393612fd4f730ff5aadc23e9b8f9648aa49ceb/mmproj-Ornith-1.5-35B-BF16.gguf + sha256: 1921a36a85aee56cd2abd27f46701802c9d85a33474792e600df6c3b282a135d +- !!merge <<: *ornith-1-5-35b-a3b + name: "ornith-1.5-35b-a3b-q6" + variants: [] + last_checked: "2026-09-14" + description: | + Ornith-1.5-35B-A3B in the Q6_K GGUF format, with the shared + BF16 vision projector for multimodal prompts. + overrides: + backend: llama-cpp + context_size: 262144 + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - vision + mmproj: llama-cpp/mmproj/ornith-1.5-35b-a3b/mmproj-BF16.gguf + options: + - use_jinja:true + parameters: + min_p: 0 + model: llama-cpp/models/ornith-1.5-35b-a3b/Ornith-1.5-35B-Q6_K.gguf + repeat_penalty: 1 + temperature: 0.6 + top_k: 20 + top_p: 0.95 + template: + use_tokenizer_template: true + files: + - filename: llama-cpp/models/ornith-1.5-35b-a3b/Ornith-1.5-35B-Q6_K.gguf + uri: https://huggingface.co/ornith-ai/Ornith-1.5-35B-A3B-GGUF/resolve/12393612fd4f730ff5aadc23e9b8f9648aa49ceb/Ornith-1.5-35B-Q6_K.gguf + sha256: 15d4658bbfc9c6034621729c15bbb50662c82b32a7ddd9624a1e545a74bdbb4b + - filename: llama-cpp/mmproj/ornith-1.5-35b-a3b/mmproj-BF16.gguf + uri: https://huggingface.co/ornith-ai/Ornith-1.5-35B-A3B-GGUF/resolve/12393612fd4f730ff5aadc23e9b8f9648aa49ceb/mmproj-Ornith-1.5-35B-BF16.gguf + sha256: 1921a36a85aee56cd2abd27f46701802c9d85a33474792e600df6c3b282a135d - &nex-n2-5-mini name: "nex-n2.5-mini-q4" variants: @@ -6439,90 +6642,6 @@ - filename: llama-cpp/models/DeepSeek-V4-Pro-0813-UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00020-of-00020.gguf uri: huggingface://unsloth/DeepSeek-V4-Pro-0813-GGUF/UD-Q4_K_XL/DeepSeek-V4-Pro-0813-UD-Q4_K_XL-00020-of-00020.gguf sha256: 5978999836fd1fe7ad7da6232a70981bd7f98274ed71d3b06652ba42c7b46031 -- name: instella-moe-16b-a3b-think - url: github:mudler/LocalAI/gallery/virtual.yaml@master - urls: - - https://huggingface.co/amd/Instella-MoE-16B-A3B-Think - - https://huggingface.co/DevQuasar/amd.Instella-MoE-16B-A3B-Think-GGUF - description: | - AMD Instella-MoE-16B-A3B-Think is a reasoning and instruction-following - mixture-of-experts model with 16 billion total parameters and 3 billion - active parameters. It supports long-form reasoning, chat, coding, and tool - use. This entry uses the Q4_K_M GGUF quantization. - license: other - tags: - - llm - - gguf - - deepseek-v3 - - moe - - reasoning - - thinking - - coding - - tool-use - - cpu - - gpu - last_checked: "2026-08-03" - variants: - - model: instella-moe-16b-a3b-think-q8 - overrides: - backend: llama-cpp - function: - automatic_tool_parsing_fallback: true - grammar: - disable: true - known_usecases: - - chat - options: - - use_jinja:true - parameters: - model: amd.Instella-MoE-16B-A3B-Think.f16.gguf.Q4_K_M.gguf - template: - use_tokenizer_template: true - files: - - filename: amd.Instella-MoE-16B-A3B-Think.f16.gguf.Q4_K_M.gguf - sha256: f776ddfee5dc6c808265c3f0a9c1b384c98058bfa551fe5df656fe712eb4c7f3 - uri: huggingface://DevQuasar/amd.Instella-MoE-16B-A3B-Think-GGUF/Q4_K_M/amd.Instella-MoE-16B-A3B-Think.f16.gguf.Q4_K_M.gguf -- name: instella-moe-16b-a3b-think-q8 - url: github:mudler/LocalAI/gallery/virtual.yaml@master - urls: - - https://huggingface.co/amd/Instella-MoE-16B-A3B-Think - - https://huggingface.co/DevQuasar/amd.Instella-MoE-16B-A3B-Think-GGUF - description: | - AMD Instella-MoE-16B-A3B-Think is a reasoning and instruction-following - mixture-of-experts model with 16 billion total parameters and 3 billion - active parameters. It supports long-form reasoning, chat, coding, and tool - use. This entry uses the near-lossless Q8_0 GGUF quantization. - license: other - tags: - - llm - - gguf - - deepseek-v3 - - moe - - reasoning - - thinking - - coding - - tool-use - - cpu - - gpu - last_checked: "2026-08-03" - overrides: - backend: llama-cpp - function: - automatic_tool_parsing_fallback: true - grammar: - disable: true - known_usecases: - - chat - options: - - use_jinja:true - parameters: - model: amd.Instella-MoE-16B-A3B-Think.f16.gguf.Q8_0.gguf - template: - use_tokenizer_template: true - files: - - filename: amd.Instella-MoE-16B-A3B-Think.f16.gguf.Q8_0.gguf - sha256: 6bff0a99d06be8ecf7d9c7f65584bfc2b24526ab0bab32dbfe7dde4237cd7c33 - uri: huggingface://DevQuasar/amd.Instella-MoE-16B-A3B-Think-GGUF/Q8_0/amd.Instella-MoE-16B-A3B-Think.f16.gguf.Q8_0.gguf - name: "parable-granite-4.1-3b-claude-fable-5" url: "github:mudler/LocalAI/gallery/virtual.yaml@master" urls: @@ -7377,6 +7496,244 @@ - filename: llama-cpp/mmproj/Hermes3.6-35B-A3B-Uncensored-Genesis-V7/mmproj-Hermes3.6-35B-A3B-Uncensored-Genesis-F16.gguf uri: huggingface://LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-Hermes-V7-GGUF/mmproj-Hermes3.6-35B-A3B-Uncensored-Genesis-F16.gguf sha256: f7197461d8581cd9be42384d4afe3f851d0844f6d72846d5c97e3dcc931b3cbc +- &qwen3-6-35b-a3b-genesis-hermes-final + name: "qwen3.6-35b-a3b-genesis-hermes-final" + url: "github:mudler/LocalAI/gallery/virtual.yaml@master" + variants: + - model: qwen3.6-35b-a3b-genesis-hermes-final-apex-compact + - model: qwen3.6-35b-a3b-genesis-hermes-final-mtp-apex + - model: qwen3.6-35b-a3b-genesis-hermes-final-mtp-apex-compact + - model: qwen3.6-35b-a3b-genesis-hermes-final-q8-k-p + urls: + - https://huggingface.co/HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive + - https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-Hermes-Final-GGUF + description: | + Qwen3.6-35B-A3B Genesis Hermes Final is a multimodal mixture-of-experts + model with 35B total parameters and about 3B active per token. This + uncensored derivative combines the HauhauCS base with Hermes + function-calling data and the author's Genesis weight processing. + This build uses APEX and includes the F16 vision projector. + license: "apache-2.0" + tags: + - llm + - gguf + - vision + - multimodal + - reasoning + - coding + - uncensored + - moe + last_checked: "2026-09-14" + overrides: + backend: llama-cpp + context_size: 131072 + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - vision + mmproj: llama-cpp/mmproj/Hermes3.6-35B-A3B-Uncensored-Genesis-Final/mmproj-Hermes3.6-35B-A3B-Uncensored-Genesis-Final-F16.gguf + options: + - use_jinja:true + parameters: + min_p: 0 + model: llama-cpp/models/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-APEX/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-APEX.gguf + repeat_penalty: 1 + temperature: 0.6 + top_k: 20 + top_p: 1 + template: + use_tokenizer_template: true + files: + - filename: llama-cpp/models/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-APEX/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-APEX.gguf + sha256: caefdcec9364e903fc7edac90c39df257f1d6671b5fc389a3cb7dc8ce8204a3b + uri: https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-Hermes-Final-GGUF/resolve/d0cf3294f07f2c422f0cf793a75fa48f61d48931/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-APEX.gguf + - filename: llama-cpp/mmproj/Hermes3.6-35B-A3B-Uncensored-Genesis-Final/mmproj-Hermes3.6-35B-A3B-Uncensored-Genesis-Final-F16.gguf + sha256: 5129bb5eb19e4346c0f2071f1ce8e1b0a076e0ab08d57a77d6033ce01235252c + uri: https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-Hermes-Final-GGUF/resolve/d0cf3294f07f2c422f0cf793a75fa48f61d48931/mmproj-Hermes3.6-35B-A3B-Uncensored-Genesis-Final-F16.gguf +- !!merge <<: *qwen3-6-35b-a3b-genesis-hermes-final + name: "qwen3.6-35b-a3b-genesis-hermes-final-apex-compact" + variants: [] + description: | + Qwen3.6-35B-A3B Genesis Hermes Final is a multimodal mixture-of-experts + model with 35B total parameters and about 3B active per token. This + uncensored derivative combines the HauhauCS base with Hermes + function-calling data and the author's Genesis weight processing. + This build uses APEX Compact and includes the F16 vision projector. + overrides: + backend: llama-cpp + context_size: 131072 + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - vision + mmproj: llama-cpp/mmproj/Hermes3.6-35B-A3B-Uncensored-Genesis-Final/mmproj-Hermes3.6-35B-A3B-Uncensored-Genesis-Final-F16.gguf + options: + - use_jinja:true + parameters: + min_p: 0 + model: llama-cpp/models/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-APEX-Compact/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-APEX-Compact.gguf + repeat_penalty: 1 + temperature: 0.6 + top_k: 20 + top_p: 1 + template: + use_tokenizer_template: true + files: + - filename: llama-cpp/models/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-APEX-Compact/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-APEX-Compact.gguf + sha256: c55f3a0acc94d28a192cf1262a9cc5dbdbe15ce7bad75894ce2b0d7e8b66ddb8 + uri: https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-Hermes-Final-GGUF/resolve/d0cf3294f07f2c422f0cf793a75fa48f61d48931/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-APEX-Compact.gguf + - filename: llama-cpp/mmproj/Hermes3.6-35B-A3B-Uncensored-Genesis-Final/mmproj-Hermes3.6-35B-A3B-Uncensored-Genesis-Final-F16.gguf + sha256: 5129bb5eb19e4346c0f2071f1ce8e1b0a076e0ab08d57a77d6033ce01235252c + uri: https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-Hermes-Final-GGUF/resolve/d0cf3294f07f2c422f0cf793a75fa48f61d48931/mmproj-Hermes3.6-35B-A3B-Uncensored-Genesis-Final-F16.gguf +- !!merge <<: *qwen3-6-35b-a3b-genesis-hermes-final + name: "qwen3.6-35b-a3b-genesis-hermes-final-mtp-apex" + variants: [] + description: | + Qwen3.6-35B-A3B Genesis Hermes Final is a multimodal mixture-of-experts + model with 35B total parameters and about 3B active per token. This + uncensored derivative combines the HauhauCS base with Hermes + function-calling data and the author's Genesis weight processing. + This build uses APEX and includes the F16 vision projector. + Native multi-token prediction is enabled for speculative decoding. + tags: + - llm + - gguf + - vision + - multimodal + - reasoning + - coding + - uncensored + - moe + - mtp + overrides: + backend: llama-cpp + context_size: 131072 + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - vision + mmproj: llama-cpp/mmproj/Hermes3.6-35B-A3B-Uncensored-Genesis-Final/mmproj-Hermes3.6-35B-A3B-Uncensored-Genesis-Final-F16.gguf + options: + - use_jinja:true + - spec_type:draft-mtp + - spec_n_max:6 + - spec_p_min:0.75 + parameters: + min_p: 0 + model: llama-cpp/models/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-MTP-APEX/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-MTP-APEX.gguf + repeat_penalty: 1 + temperature: 0.6 + top_k: 20 + top_p: 1 + template: + use_tokenizer_template: true + files: + - filename: llama-cpp/models/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-MTP-APEX/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-MTP-APEX.gguf + sha256: 01e16a28774e4f075a2bfd92ff889cb7b004e37511e68399caf39a3f4b195722 + uri: https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-Hermes-Final-GGUF/resolve/d0cf3294f07f2c422f0cf793a75fa48f61d48931/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-MTP-APEX.gguf + - filename: llama-cpp/mmproj/Hermes3.6-35B-A3B-Uncensored-Genesis-Final/mmproj-Hermes3.6-35B-A3B-Uncensored-Genesis-Final-F16.gguf + sha256: 5129bb5eb19e4346c0f2071f1ce8e1b0a076e0ab08d57a77d6033ce01235252c + uri: https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-Hermes-Final-GGUF/resolve/d0cf3294f07f2c422f0cf793a75fa48f61d48931/mmproj-Hermes3.6-35B-A3B-Uncensored-Genesis-Final-F16.gguf +- !!merge <<: *qwen3-6-35b-a3b-genesis-hermes-final + name: "qwen3.6-35b-a3b-genesis-hermes-final-mtp-apex-compact" + variants: [] + description: | + Qwen3.6-35B-A3B Genesis Hermes Final is a multimodal mixture-of-experts + model with 35B total parameters and about 3B active per token. This + uncensored derivative combines the HauhauCS base with Hermes + function-calling data and the author's Genesis weight processing. + This build uses APEX Compact and includes the F16 vision projector. + Native multi-token prediction is enabled for speculative decoding. + tags: + - llm + - gguf + - vision + - multimodal + - reasoning + - coding + - uncensored + - moe + - mtp + overrides: + backend: llama-cpp + context_size: 131072 + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - vision + mmproj: llama-cpp/mmproj/Hermes3.6-35B-A3B-Uncensored-Genesis-Final/mmproj-Hermes3.6-35B-A3B-Uncensored-Genesis-Final-F16.gguf + options: + - use_jinja:true + - spec_type:draft-mtp + - spec_n_max:6 + - spec_p_min:0.75 + parameters: + min_p: 0 + model: llama-cpp/models/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-MTP-APEX-Compact/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-MTP-APEX-Compact.gguf + repeat_penalty: 1 + temperature: 0.6 + top_k: 20 + top_p: 1 + template: + use_tokenizer_template: true + files: + - filename: llama-cpp/models/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-MTP-APEX-Compact/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-MTP-APEX-Compact.gguf + sha256: fe921f9f7d3f48f9fbda1f28bb3e96824db051958cbc6e44dd0e5bebc690829a + uri: https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-Hermes-Final-GGUF/resolve/d0cf3294f07f2c422f0cf793a75fa48f61d48931/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-MTP-APEX-Compact.gguf + - filename: llama-cpp/mmproj/Hermes3.6-35B-A3B-Uncensored-Genesis-Final/mmproj-Hermes3.6-35B-A3B-Uncensored-Genesis-Final-F16.gguf + sha256: 5129bb5eb19e4346c0f2071f1ce8e1b0a076e0ab08d57a77d6033ce01235252c + uri: https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-Hermes-Final-GGUF/resolve/d0cf3294f07f2c422f0cf793a75fa48f61d48931/mmproj-Hermes3.6-35B-A3B-Uncensored-Genesis-Final-F16.gguf +- !!merge <<: *qwen3-6-35b-a3b-genesis-hermes-final + name: "qwen3.6-35b-a3b-genesis-hermes-final-q8-k-p" + variants: [] + description: | + Qwen3.6-35B-A3B Genesis Hermes Final is a multimodal mixture-of-experts + model with 35B total parameters and about 3B active per token. This + uncensored derivative combines the HauhauCS base with Hermes + function-calling data and the author's Genesis weight processing. + This build uses Q8_K_P and includes the F16 vision projector. + overrides: + backend: llama-cpp + context_size: 131072 + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - vision + mmproj: llama-cpp/mmproj/Hermes3.6-35B-A3B-Uncensored-Genesis-Final/mmproj-Hermes3.6-35B-A3B-Uncensored-Genesis-Final-F16.gguf + options: + - use_jinja:true + parameters: + min_p: 0 + model: llama-cpp/models/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-Q8_K_P/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-Q8_K_P.gguf + repeat_penalty: 1 + temperature: 0.6 + top_k: 20 + top_p: 1 + template: + use_tokenizer_template: true + files: + - filename: llama-cpp/models/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-Q8_K_P/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-Q8_K_P.gguf + sha256: e1ac7246dd3ebfb036732874dc57acbacc8b0d7d02df4417e2ac0d1f376e8d1d + uri: https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-Hermes-Final-GGUF/resolve/d0cf3294f07f2c422f0cf793a75fa48f61d48931/Hermes3.6-35B-A3B-Uncensored-Genesis-Final-Q8_K_P.gguf + - filename: llama-cpp/mmproj/Hermes3.6-35B-A3B-Uncensored-Genesis-Final/mmproj-Hermes3.6-35B-A3B-Uncensored-Genesis-Final-F16.gguf + sha256: 5129bb5eb19e4346c0f2071f1ce8e1b0a076e0ab08d57a77d6033ce01235252c + uri: https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-Hermes-Final-GGUF/resolve/d0cf3294f07f2c422f0cf793a75fa48f61d48931/mmproj-Hermes3.6-35B-A3B-Uncensored-Genesis-Final-F16.gguf + - &kat-coder-v2-5-dev name: "kat-coder-v2.5-dev" variants: @@ -11203,6 +11560,88 @@ - filename: llama-cpp/mmproj/Qwopus3.6-27B-Coder-MTP-GGUF/mmproj-F32.gguf sha256: 32f7ea0600c07272547da401d460f8abbd980f3a57b69d6df87be0e2505e0b9c uri: https://huggingface.co/Jackrong/Qwopus3.6-27B-Coder-MTP-GGUF/resolve/main/mmproj-F32.gguf +- &gemma-4-31b-ortenzya + name: gemma-4-31b-ortenzya-q4 + variants: + - model: gemma-4-31b-ortenzya-q8 + url: github:mudler/LocalAI/gallery/virtual.yaml@master + urls: + - https://huggingface.co/llmfan46/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic + - https://huggingface.co/llmfan46/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-GGUF + license: apache-2.0 + icon: https://ai.google.dev/gemma/images/gemma4_banner.png + tags: + - llm + - gguf + - gemma4 + - multimodal + - reasoning + - creative-writing + description: | + Ortenzya is a Gemma 4 31B instruction-tuned model from llmfan46 for + creative writing, translation, and roleplay. This entry uses the Q4_K_M + GGUF with the BF16 vision projector and a 32,768-token context. + overrides: + backend: llama-cpp + context_size: 32768 + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - vision + mmproj: llama-cpp/mmproj/gemma-4-31b-ortenzya/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-mmproj-BF16.gguf + options: + - use_jinja:true + parameters: + model: llama-cpp/models/gemma-4-31b-ortenzya/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-Q4_K_M.gguf + temperature: 1 + top_p: 0.95 + top_k: 64 + template: + use_tokenizer_template: true + files: + - filename: llama-cpp/models/gemma-4-31b-ortenzya/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-Q4_K_M.gguf + uri: https://huggingface.co/llmfan46/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-GGUF/resolve/d7c96e816759b50ba7d0afe3a9e36deb7b096f16/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-Q4_K_M.gguf + sha256: aefb0b5a93362b8880e2639c15ea807220351c7abfa8e0f6903ca25f31844981 + - filename: llama-cpp/mmproj/gemma-4-31b-ortenzya/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-mmproj-BF16.gguf + uri: https://huggingface.co/llmfan46/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-GGUF/resolve/d7c96e816759b50ba7d0afe3a9e36deb7b096f16/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-mmproj-BF16.gguf + sha256: c040f82999134f7ba8dddf1c9e3451822c8ae9af51b60a01619df43a2a3a7128 +- !!merge <<: *gemma-4-31b-ortenzya + name: gemma-4-31b-ortenzya-q8 + variants: [] + description: | + Ortenzya is a Gemma 4 31B instruction-tuned model from llmfan46 for + creative writing, translation, and roleplay. This entry uses the Q8_0 + GGUF with the BF16 vision projector and a 32,768-token context. + overrides: + backend: llama-cpp + context_size: 32768 + function: + automatic_tool_parsing_fallback: true + grammar: + disable: true + known_usecases: + - chat + - vision + mmproj: llama-cpp/mmproj/gemma-4-31b-ortenzya/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-mmproj-BF16.gguf + options: + - use_jinja:true + parameters: + model: llama-cpp/models/gemma-4-31b-ortenzya/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-Q8_0.gguf + temperature: 1 + top_p: 0.95 + top_k: 64 + template: + use_tokenizer_template: true + files: + - filename: llama-cpp/models/gemma-4-31b-ortenzya/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-Q8_0.gguf + uri: https://huggingface.co/llmfan46/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-GGUF/resolve/d7c96e816759b50ba7d0afe3a9e36deb7b096f16/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-Q8_0.gguf + sha256: 688170f6d6e03edc33ac86891eb3c37ec6cb1739f5db8625a84a1e4cdba17586 + - filename: llama-cpp/mmproj/gemma-4-31b-ortenzya/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-mmproj-BF16.gguf + uri: https://huggingface.co/llmfan46/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-GGUF/resolve/d7c96e816759b50ba7d0afe3a9e36deb7b096f16/gemma-4-Ortenzya-The-Creative-Wordsmith-31B-it-uncensored-heretic-mmproj-BF16.gguf + sha256: c040f82999134f7ba8dddf1c9e3451822c8ae9af51b60a01619df43a2a3a7128 - &gemma-4-31b-scotoma-2 name: "gemma-4-31b-scotoma-2-q4" variants: @@ -14597,6 +15036,107 @@ - filename: llama-cpp/mmproj/mmproj-Qwen_Qwen3.5-2B-f16.gguf sha256: 044a0ea136cca70711ae16e23b24d754b44eab6f2462d187aee4d7c7a9503d36 uri: https://huggingface.co/bartowski/Qwen_Qwen3.5-2B-GGUF/resolve/main/mmproj-Qwen_Qwen3.5-2B-f16.gguf +- &neohorse-1-4b + name: neohorse-1-4b-q4 + url: github:mudler/LocalAI/gallery/virtual.yaml@master + urls: + - https://huggingface.co/TokenRhythm/NeoHorse-1-4B + - https://huggingface.co/mradermacher/NeoHorse-1-4B-GGUF + license: apache-2.0 + tags: + - llm + - gguf + - cpu + - gpu + - reasoning + - coding + variants: + - model: neohorse-1-4b-q6 + - model: neohorse-1-4b-q8 + description: | + NeoHorse-1-4B is TokenRhythm's text-only Qwen3.5-4B fine-tune for coding, + reasoning, and agentic tasks. This build uses Q4_K_M GGUF weights with + the embedded Jinja chat template and a 32K-token default context. + overrides: + backend: llama-cpp + context_size: 32768 + known_usecases: + - chat + options: + - use_jinja:true + parameters: + model: NeoHorse-1-4B.Q4_K_M.gguf + temperature: 1 + top_p: 0.95 + top_k: 20 + min_p: 0 + presence_penalty: 1.5 + repeat_penalty: 1 + template: + use_tokenizer_template: true + files: + - filename: NeoHorse-1-4B.Q4_K_M.gguf + uri: https://huggingface.co/mradermacher/NeoHorse-1-4B-GGUF/resolve/926d357de701ecf787324228734866faaf02054e/NeoHorse-1-4B.Q4_K_M.gguf + sha256: 9a8426d0a5a1035d7e35ccc02794c404e027cca8e51f936b225d26d93bc44753 + +- !!merge <<: *neohorse-1-4b + name: neohorse-1-4b-q6 + variants: [] + description: | + NeoHorse-1-4B is TokenRhythm's text-only Qwen3.5-4B fine-tune for coding, + reasoning, and agentic tasks. This build uses Q6_K GGUF weights with + the embedded Jinja chat template and a 32K-token default context. + overrides: + backend: llama-cpp + context_size: 32768 + known_usecases: + - chat + options: + - use_jinja:true + parameters: + model: NeoHorse-1-4B.Q6_K.gguf + temperature: 1 + top_p: 0.95 + top_k: 20 + min_p: 0 + presence_penalty: 1.5 + repeat_penalty: 1 + template: + use_tokenizer_template: true + files: + - filename: NeoHorse-1-4B.Q6_K.gguf + uri: https://huggingface.co/mradermacher/NeoHorse-1-4B-GGUF/resolve/926d357de701ecf787324228734866faaf02054e/NeoHorse-1-4B.Q6_K.gguf + sha256: d29d06d4b778b21a4b7145428ba5d709bf368fc6dc7321e0af3463c1718fec41 + +- !!merge <<: *neohorse-1-4b + name: neohorse-1-4b-q8 + variants: [] + description: | + NeoHorse-1-4B is TokenRhythm's text-only Qwen3.5-4B fine-tune for coding, + reasoning, and agentic tasks. This build uses Q8_0 GGUF weights with + the embedded Jinja chat template and a 32K-token default context. + overrides: + backend: llama-cpp + context_size: 32768 + known_usecases: + - chat + options: + - use_jinja:true + parameters: + model: NeoHorse-1-4B.Q8_0.gguf + temperature: 1 + top_p: 0.95 + top_k: 20 + min_p: 0 + presence_penalty: 1.5 + repeat_penalty: 1 + template: + use_tokenizer_template: true + files: + - filename: NeoHorse-1-4B.Q8_0.gguf + uri: https://huggingface.co/mradermacher/NeoHorse-1-4B-GGUF/resolve/926d357de701ecf787324228734866faaf02054e/NeoHorse-1-4B.Q8_0.gguf + sha256: ec5e02b4d3a102335333a692b1b83ee454e249e06a6d27f00b2f5bc81a199c98 + - name: qwen_qwen3.5-4b url: github:mudler/LocalAI/gallery/virtual.yaml@master urls: @@ -27205,6 +27745,7 @@ Q4_K_M GGUF quantization and the model's embedded chat template. variants: - model: minicpm5-2b-q8 + - model: minicpm5-2b-f16 overrides: backend: llama-cpp known_usecases: @@ -27246,6 +27787,30 @@ - filename: llama-cpp/models/minicpm5-2b/MiniCPM5-2B-Q8_0.gguf uri: huggingface://openbmb/MiniCPM5-2B-GGUF/MiniCPM5-2B-Q8_0.gguf sha256: c5415f8989bf88a8288f1b55a3cc371af53c07b0faa220a63bd7a990cfaba078 +- !!merge <<: *minicpm5-2b + name: minicpm5-2b-f16 + description: | + MiniCPM5-2B is OpenBMB's compact Llama-based language model for English + and Chinese chat, coding, and reasoning. This entry uses the official + F16 GGUF build and the model's embedded chat template. + variants: null + overrides: + backend: llama-cpp + context_size: 8192 + known_usecases: + - chat + options: + - use_jinja:true + template: + use_tokenizer_template: true + parameters: + model: llama-cpp/models/minicpm5-2b/MiniCPM5-2B-F16.gguf + temperature: 1.0 + top_p: 0.95 + files: + - filename: llama-cpp/models/minicpm5-2b/MiniCPM5-2B-F16.gguf + uri: https://huggingface.co/openbmb/MiniCPM5-2B-GGUF/resolve/2079a22f3beaa4e306449978533478fe0522f4b3/MiniCPM5-2B-F16.gguf + sha256: 0ffba3682a853295566b98bc38c0ab755d6b2d91b994bc031ed727a08cfcdf25 - &minicpm5-1b name: minicpm5-1b url: github:mudler/LocalAI/gallery/chatml.yaml@master @@ -60225,7 +60790,7 @@ model: audio-cpp/sortformer-diar-4spk-v1-q8_0.gguf files: - filename: audio-cpp/sortformer-diar-4spk-v1-q8_0.gguf - sha256: 84744b047296daf2177aed9dd2735497152ff83f49ec170d2a6a18cbcb93ef19 + sha256: 4fa6a3e30c4a1c6cc1da455268806edc93432ca1e1b5b6923e942be8e6e479ff uri: huggingface://audio-cpp/audio.cpp-gguf/Sortformer-Diar-4spk-v1-GGUF/sortformer-diar-4spk-v1-q8_0.gguf - name: audio-cpp-htdemucs url: github:mudler/LocalAI/gallery/virtual.yaml@master