THU-KEG/LongWriter-V-72B

Full fine-tune vs Qwen/Qwen2.5-VL-72B-Instruct. Fine-tune: 97% of the weights changed, across 80 of 80 layers.

Tensors changed: Embeddings 0/2, Norms 124/226, Attention 559/688, MLP 240/436, lm_head 1/1

2 derivatives checked

• mradermacher/LongWriter-V-72B-i1-GGUF: Quantization
• mradermacher/LongWriter-V-72B-GGUF: Quantization

On the Hub · Downloads on Model Pulse