GreenBitAI/GLM-5.3-Flash-4bit-paged

Same weights, repacked vs pipenetwork/GLM-5.3-Flash-MLX-4bit. Same weights in different files (re-saved, re-sharded or new metadata).

Tensors changed: lm_head 0/3, Embeddings 0/71, Attention 0/1504, MLP 0/771, Norms 0/268, other 0/3

On the Hub · Downloads on Model Pulse