EZCon/Huihui-Qwen3-VL-2B-Instruct-abliterated-mlx

Same weights, repacked vs huihui-ai/Huihui-Qwen3-VL-2B-Instruct-abliterated. Same weights in different files (re-saved, re-sharded or new metadata), plus 310 new tensors.

Tensors changed: Attention 0/96, MLP 0/112, Norms 0/104, Embeddings 0/2

On the Hub · Downloads on Model Pulse