kaitchup/MiniMax-M3-GGUF-MoQ

Same weights, converted vs MiniMaxAI/MiniMax-M3. Same weights stored in another dtype (BF16/F32 → IQ2_XS/IQ1_S), plus 228 new tensors.

Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/241, MLP 0/180, Attention 0/240, other 0/57

On the Hub · Downloads on Model Pulse