xumingsheng/dunhuang-qwen2.5-vl-7b

Dunhuang Fine-tuned Qwen2.5-VL-7B

This model is fine-tuned from Qwen2.5-VL-7B-Instruct on 51,570 Dunhuang murals image-text pairs.

Training Details

• Base Model: Qwen2.5-VL-7B-Instruct
• Training samples: 51,570
• Hardware: 4x NVIDIA H200
• Training time: 2.8 hours
• Epochs: 3

Features

• Classical Chinese language style (文言文风格)
• Dunhuang art understanding
• Buddhist cultural context