DeepCybo/PhysBrain1.5-8B

Full fine-tune vs Qwen/Qwen3-VL-8B-Instruct. Fine-tune: 92% of the weights changed, across 36 of 36 layers.

Tensors changed: Norms 138/261, MLP 108/232, Attention 144/252, Embeddings 0/3

2 derivatives checked

• mradermacher/PhysBrain1.5-8B-i1-GGUF: Quantization (fewer layers)
• mradermacher/PhysBrain1.5-8B-GGUF: Same weights, converted (fewer layers)

On the Hub · Downloads on Model Pulse