itod/mistral-7B-instruct-v0.2-f16

• Source Mistral 7B model:
  https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2/
• This model is converted from Bfloat16 datatype to Float16 datatype with convert tool from:
  https://github.com/ggerganov/llama.cpp
• Deployment on CUDA GPU:
  Pull the ready-made llama.cpp container:

docker pull ghcr.io/ggerganov/llama.cpp:server-cuda

Assuming mistral-7B-instruct-v0.2-fp16.gguf file is downloaded to /path/to/models directory on the local machine, run the container accesing the model with:

docker run --gpus all -v /path/to/models:/models -p 8000:8000 ghcr.io/ggerganov/llama.cpp:server-cuda -m /models/mistral-7B-instruct-v0.2-fp16.gguf --port 8000 --host 0.0.0.0 -n 512 --n-gpu-layers 50

• Test the deployment accessing the model with the browser at http://localhost:8000
• llama.cpp server also provides OpenAI compatible API
• If CUDA GPU is not available, the version of the model converted to int8 may be interesting, available in this repo:
  https://huggingface.co/itod/mistral-7B-instruct-v0.2-q8
• More details about usage is avalable in llama.cpp documentation:
  https://github.com/ggerganov/llama.cpp/tree/master/examples/server