NorwAI

/

NorwAI-Mistral-7B-instruct-overfitting

like 17

Same as with the regular Mistral and Llama-models, this PR will add quantized GGUF versions of the model s.t. you can run it on your own machine (for instance using Ollama). No particular config choices were made, other than using f16 vectors (matches previous, as well as the safetensors files).

Ready on my end!