elbelga/Huihui-Qwen3.5-35B-A3B-Claude-4.6-Opus-abliterated_MXFP4_MOE

Same weights, converted vs huihui-ai/Huihui-Qwen3.5-35B-A3B-Claude-4.6-Opus-abliterated. Same weights stored in another dtype (BF16/F32 → MXFP4/BF16), plus 150 new tensors.

Tensors changed: lm_head 0/1, Embeddings 0/1, Attention 0/40, Norms 0/101, MLP 0/160

On the Hub · Downloads on Model Pulse