BatsResearch

/

Llama-3.1-8B-bonito-v1

like 4

Thanks for posting the issue. A community user has created an ollama version of Bonito-v1 (with Mistral as the backbone). I will create one with the llama-3.1 checkpoint if there is more interest.

I'm interested. Do you think it could be useful to have a longer text version?

I’m interested in using a GGUF file with 4-bit quantization on this model with my RTX 4060 Ti (16GB).