Marcin116/Qwen3.8-27B-Uncensored-W4A16-RTX3090-MTP4-embed-int4

Same weights, repacked vs noon-at-cgn/Qwen3.8-27B-Uncensored-W4A16-AutoRound. Same weights in different files (re-saved, re-sharded or new metadata), plus 33 new tensors.

Tensors changed: Attention 0/972, MLP 0/688, Norms 0/325, Embeddings 0/4

On the Hub · Downloads on Model Pulse