albucino/Qwen3.8-Flash-Next-W4A16-FP8PLE

Full fine-tune vs Intel/Qwen3.8-Flash-Next-W4A16-AutoRound. Fine-tune: 90% of the weights changed, across 1 of 48 layers.

Tensors changed: Embeddings 128/135, MLP 0/1925, Attention 0/600, Norms 0/294, other 0/5, lm_head 0/1

On the Hub · Downloads on Model Pulse