Fine-tuned from Qwen/Qwen3-VL-8B-Instruct. Same architecture, tokenizer, processor and chat template as the base model.
Qwen/Qwen3-VL-8B-Instruct
Chat template
Files info
Base model