sparkarena/Minimax-M3-v0-NVFP4-REAP25

Partial fine-tune vs sparkarena/Minimax-M3-v0-NVFP4. Partial change: 11% of the weights changed (mlp), in 57 of 60 layers.

Tensors changed: MLP 362/1355, Attention 0/1554, lm_head 0/1, Embeddings 0/2, Norms 0/455, other 0/4

On the Hub · Downloads on Model Pulse