Vision-Language Models (VLMs)

• Qwen-VL-Chat-1.1: 🤗, 🗄️
• TextMonkey: An OCR-Free Large Multimodal Model for Understanding Document
• LLaVA-NeXT (LLaVA-1.6)
• MoE-LLaVA: Mixture of Experts for Large Vision-Language Models 🤗
• DeepSeek-VL-7B-Chat: 🤗
• DocOwl 1.5: Unified Structure Learning for OCR-free Document Understanding, sum
• UReader: Universal OCR-free Visually-situated Language Understanding
• MME: A Comprehensive Evaluation Benchmark for Multimodal LLM
• VILA-2.7b: Vision Model from NVIDIA and MIT, demo, 🤗