GUIAgent/MagicGUI_RFT

Full fine-tune vs Qwen/Qwen2-VL-7B-Instruct. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

Tensors changed: Embeddings 2/2, Norms 187/187, MLP 216/216, Attention 324/324, lm_head 1/1

2 derivatives checked

• mradermacher/MagicGUI_RFT-i1-GGUF: Quantization (fewer layers)
• mradermacher/MagicGUI_RFT-GGUF: Same weights, converted (fewer layers)

On the Hub · Downloads on Model Pulse