unsloth/Qwen3.5-27B-GGUF

Same weights, converted vs Qwen/Qwen3.5-27B. Same weights stored in another dtype (BF16/F32 → BF16/F32), plus 48 new tensors.

Tensors changed: lm_head 0/1, Embeddings 0/1, MLP 0/192, Attention 0/64, Norms 0/161

1 derivatives checked

• Biomanticus/Biomanticus-ClaudeOpus4.6-Qwen3.5-27B_finetuned-gguf: Partial fine-tune

On the Hub · Downloads on Model Pulse