New model idea
#3
by
Enderchef
- opened
Could you distill MiniMaxAI/MiniMax-M1-80k onto a small, open source model like Mistral Small, QwQ, Qwen 72B, ect.
New SOTA models come out every week we can't keep track of what to distill anymore!
Haha, that's true! If you had time and leisure, which (or what kind of) model would you like to distill or create?