Consider MiniMax-M1 as a new model

The recently released MiniMax-M1 model looks interesting. With native 1 million Context Window and 80K output tokens, I think this is an interesting model to check out.

https://huggingface.co/MiniMaxAI/MiniMax-M1-80k

Performance is comparable to the new Deepseek-R1 and better than Qwen3-235B.