r0b0tlab/Qwen3.8-Flash-Next-EXL3-2.50bpw

Same weights, converted vs Qwen/Qwen3.8-Flash-Next. Same weights stored in another dtype (BF16/I64 → I16/F16), plus 303085 new tensors.

Tensors changed: Norms 0/302, other 0/7, Attention 0/357, MLP 0/274, Embeddings 0/4

On the Hub · Downloads on Model Pulse