MirroS-Lab/Code-as-World-VL-4B

Partial fine-tune vs Qwen/Qwen3.5-4B. Partial change: 86% of the weights changed (attention, embeddings, mlp, norms), in 32 of 32 layers.

Tensors changed: Embeddings 3/4, MLP 172/196, Attention 271/320, Norms 99/203

1 derivatives checked

• prithivMLmods/Code-as-World-VL-4B-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse