Vision Understanding on RealworldQA
75.4Overall ScoreGPT-4o-0513
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4o-05132024.09 | 75.4 | |
| OneVision2024.09 | 69.9 | |
| Qwen2.5-VLSize=7.0B, Capability=Und. only2025.05 | 68.9 | |
| MetaMorphSize=8.0B, Capability=Und. and Gen.2025.05 | 68.8 | |
| MindOmniSize=7.0B, Capability=Und. and Gen.2025.05 | 68.1 | |
| IXC-2.52024.09 | 67.8 | |
| POINTS-7BLLM=Qwen-2.5-7B2024.09 | 67.3 | |
| Cambrian-34B2024.09 | 67.1 | |
| LLaVA-OVSize=7.0B, Capability=Und. only2025.05 | 66.8 | |
| POINTS-9BLLM=Yi-1.5-9B2024.09 | 65.9 | |
| Ovis1.5-LLAMA3-8B2024.09 | 64.2 | |
| InternVL2-8B2024.09 | 64.2 | |
| Gemini-1.5-Pro2024.09 | 64.1 | |
| Idefics3-LLAMA3-8B2024.09 | 62.6 | |
| Claude3.5-Sonnet2024.09 | 60.1 | |
| TokenFlow-XLSize=14B, Capability=Und. and Gen.2025.05 | 58.3 | |
| Emu3-ChatSize=8.0B, Capability=Und. only2025.05 | 57.4 |