zmzfpc/crane-next-80b

Full fine-tune vs Qwen/Qwen3-Next-80B-A3B-Instruct. Fine-tune: 93% of the weights changed, across 48 of 48 layers.

Tensors changed: Embeddings 0/2, Norms 115/163, Attention 205/268, MLP 2548/2565, lm_head 1/1, other 1/1

2 derivatives checked

• mradermacher/crane-next-80b-GGUF: Quantization
• mradermacher/crane-next-80b-i1-GGUF: Quantization

On the Hub · Downloads on Model Pulse