diff --git a/gallery/index.yaml b/gallery/index.yaml index 6662f78bf8a6..a0b70498ea52 100644 --- a/gallery/index.yaml +++ b/gallery/index.yaml @@ -1,4 +1,33 @@ --- +- name: "editthinker-8b" + url: "github:mudler/LocalAI/gallery/virtual.yaml@master" + urls: + - https://huggingface.co/mradermacher/EditThinker-8B-GGUF + description: | + The model, **EditThinker-8B (quantized version)**, is a large language model designed for tasks like text generation, reasoning, and multi-turn dialogue. It is a compressed version of the original EditThinker-8B model, optimized for efficiency with quantization techniques (e.g., Q4_K_S, Q2_K, etc.). It retains the core capabilities of the original model but is tailored for deployment in resource-constrained environments. The quantized version is suitable for applications requiring reduced memory usage and faster inference. For the original, unquantized model, refer to the official Hugging Face repository. + overrides: + parameters: + model: llama-cpp/models/EditThinker-8B.Q4_K_M.gguf + name: EditThinker-8B-GGUF + backend: llama-cpp + template: + use_tokenizer_template: true + known_usecases: + - chat + function: + grammar: + disable: true + mmproj: llama-cpp/mmproj/EditThinker-8B.mmproj-f16.gguf + description: Imported from https://huggingface.co/mradermacher/EditThinker-8B-GGUF + options: + - use_jinja:true + files: + - filename: llama-cpp/models/EditThinker-8B.Q4_K_M.gguf + sha256: 815a2480340b91200f0d8c3ca3d986bb70d2b6329c9ed9572b98c8bb560ecd57 + uri: https://huggingface.co/mradermacher/EditThinker-8B-GGUF/resolve/main/EditThinker-8B.Q4_K_M.gguf + - filename: llama-cpp/mmproj/EditThinker-8B.mmproj-f16.gguf + sha256: c18b616c54f24dd1dec11bcda81bf0f58f49f2776ea4ed77abc5f5e365fbdef8 + uri: https://huggingface.co/mradermacher/EditThinker-8B-GGUF/resolve/main/EditThinker-8B.mmproj-f16.gguf - name: "lfm2.5-1.2b-nova-function-calling" url: "github:mudler/LocalAI/gallery/virtual.yaml@master" urls: