Document Visual Question Answering on InfographicVQA
88.1ANLSMiMo-VL-7B-RL
Evaluation results
| Method | Links | |
|---|---|---|
| MiMo-VL-7B-RLOpen-source=true, Architectural Paradigm=Reasoning-Enhanced OS Models2026.03 | 88.1 | |
| ORCA (Qwen3VL-8B)Open-source=true, Architectural Paradigm=ORCA (Multi-Agent Framework)2026.03 | 88 | |
| ORCA (Qwen2.5-VL-7B)Open-source=true, Architectural Paradigm=ORCA (Multi-Agent Framework)2026.03 | 86.9 | |
| ORCA (Qwen3VL-4B)Open-source=true, Architectural Paradigm=ORCA (Multi-Agent Framework)2026.03 | 85.4 | |
| Qwen2-VLOpen-source=true, Architectural Paradigm=General-Purpose VLMs2026.03 | 84.7 | |
| InternVL2-ProOpen-source=true, Architectural Paradigm=General-Purpose VLMs2026.03 | 83.3 | |
| Qwen3VL-8B-InstructOpen-source=true, Architectural Paradigm=Baseline VLMs (Single-Model)2026.03 | 83.1 | |
| GPT-4oOpen-source=false, Architectural Paradigm=General-Purpose VLMs2026.03 | 82.1 | |
| Molmo-72BOpen-source=true, Architectural Paradigm=General-Purpose VLMs2026.03 | 81.9 | |
| Qwen3VL-4B-InstructOpen-source=true, Architectural Paradigm=Baseline VLMs (Single-Model)2026.03 | 80.3 | |
| VideoLLaMA3-7BOpen-source=true, Architectural Paradigm=Reasoning-Enhanced OS Models2026.03 | 78.9 | |
| LLaVA-One-Vision-8BOpen-source=true, Architectural Paradigm=General-Purpose VLMs2026.03 | 78.4 | |
| DeepSeek-VL2Open-source=true, Architectural Paradigm=General-Purpose VLMs2026.03 | 78.1 | |
| Qwen2.5-VL-7B-InstructOpen-source=true, Architectural Paradigm=Baseline VLMs (Single-Model)2026.03 | 77.7 | |
| Qwen-VL-Max (single)Open-source=false, Architectural Paradigm=General-Purpose VLMs2026.03 | 73.4 | |
| Gemini Pro 1.5Open-source=false, Architectural Paradigm=General-Purpose VLMs2026.03 | 72.7 | |
| GPT-4 Turbo + TextractOpen-source=false, Architectural Paradigm=General-Purpose VLMs2026.03 | 71.9 | |
| Gemma-3-27B-ITOpen-source=true, Architectural Paradigm=Reasoning-Enhanced OS Models2026.03 | 70.6 | |
| Claude-3.7 SonnetOpen-source=false, Architectural Paradigm=General-Purpose VLMs2026.03 | 65.5 | |
| DocOwl-1.5Open-source=true, Architectural Paradigm=Document Understanding Models2026.03 | 50.7 | |
| Text-MonkeyOpen-source=true, Architectural Paradigm=Document Understanding Models2026.03 | 28.6 | |
| LayoutLMv2 LARGEOpen-source=true, Architectural Paradigm=Document Understanding Models2026.03 | 28.3 |