*Fine-Tuning QWEN2.5VL on a Custom Dataset* 🔥🔥🔥 DeepSeek is getting all the buzz, but, Alibaba's Qwen team dropped QWEN2.5VL, an open-source vision-language model that surpasses OpenAI's GPT-4, Anthropic's Claude 3.5 Sonnet, and Google's Gemini 2.0 Flash in tasks like video understanding, math reasoning, document analysis, and question answering. I stayed up all night 😤 to prepare a guide on how to fine-tune QWEN2.5VL on your custom dataset. I’m also working on adding QWEN2.5VL support to Maestro. Stay tuned! ⮑ 🔗 notebook: https://github.com/roboflow/notebooks/blob/main/notebooks/how-to-finetune-qwen2-5-vl-for-json-data-extraction.ipynb Credits: https://whatsapp.com/channel/0029VayIXpnKLaHhzg4Cvp12/113