MBZUAI/MediX-R1-30B

Full fine-tune vs Qwen/Qwen3-VL-30B-A3B-Instruct. Fine-tune: 100% of the weights changed, across 48 of 48 layers.

Tensors changed: Embeddings 4/4, Norms 270/309, MLP 262/268, Attention 300/300, lm_head 1/1

3 derivatives checked

• MBZUAI/MediX-R1-30B-GGUF: Quantization
• mradermacher/MediX-R1-30B-i1-GGUF: Quantization
• mradermacher/MediX-R1-30B-GGUF: Quantization

On the Hub · Downloads on Model Pulse