InfiX-ai/InfiGUI-G1-3B
Full fine-tune vs Qwen/Qwen2.5-VL-3B-Instruct. Fine-tune: 100% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 2/2, Norms 74/138, MLP 274/304, Attention 379/380
4 derivatives checked
- mradermacher/InfiGUI-G1-3B-GGUF: Same weights, converted (fewer layers)
- mradermacher/InfiGUI-G1-3B-i1-GGUF: Quantization (fewer layers)
- DevQuasar/InfiX-ai.InfiGUI-G1-3B-GGUF: Same weights, converted (fewer layers)
- OfficerChul/InfiGUI-G1-3B-Android-Control-5a: Full fine-tune