Saren/Qwen3.8-Flash-Next-W4A16-AutoRound-hybrid-MTP_int4RTN

Same weights, repacked vs Saren/Qwen3.8-Flash-Next-W4A16-AutoRound-hybrid. Same weights in different files (re-saved, re-sharded or new metadata), plus 4608 new tensors.

Tensors changed: Attention 0/764, MLP 0/1889, Norms 0/302, Embeddings 0/9, other 0/8, lm_head 0/3

1 derivatives checked

• sayyidfareed/Qwen3.8-Flash-Next-Code-Turbo-Spark: Identical copy

On the Hub · Downloads on Model Pulse