madrisight/madrimed1.2-VL-2B

Partial fine-tune vs Qwen/Qwen3-VL-2B-Thinking. Partial change: 85% of the weights changed (attention, embeddings, mlp), in 28 of 28 layers.

Tensors changed: Embeddings 1/4, Norms 0/217, MLP 140/196, Attention 160/208

2 derivatives checked

• madrisight/madrimed1.2-VL-2B-GGUF: Same weights, converted (fewer layers)
• mradermacher/madrimed1.2-VL-2B-GGUF: Same weights, converted (fewer layers)

On the Hub · Downloads on Model Pulse