SamPurkis/gpt-oss-puzzle-88B-GGUF

Same weights, converted vs nvidia/gpt-oss-puzzle-88B. Same weights stored in another dtype (U8/BF16 → MXFP4/F16), plus 252 new tensors.

Tensors changed: Norms 0/73, MLP 0/72, Attention 0/288, lm_head 0/1, Embeddings 0/1

On the Hub · Downloads on Model Pulse