mradermacher/Qwen-3.5-10.5B-Frankenmerge-Opus-4.6-Distill-GGUF

Same weights, converted vs JackBinary/Qwen-3.5-10B-Frankenmerge-Opus-4.6-Distill. Same weights stored in another dtype (BF16 → F16/F32), plus 27 new tensors.

Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/91, MLP 0/108, Attention 0/36

On the Hub · Downloads on Model Pulse