Launch80/Qwen3.8-27B-GPTQ-Int4-baked

Same weights, repacked vs SergiioB/Qwen3.8-27B-GPTQ-Int4-sym-G128-MTP-BF16. Same weights in different files (re-saved, re-sharded or new metadata), plus 36 new tensors.

Tensors changed: Embeddings 0/5, Norms 0/325, Attention 0/1180, MLP 0/880

On the Hub · Downloads on Model Pulse