模型排行
| 排名 | 模型 | 机构 | 分数 | 置信区间 | 开放状态 | 证据 |
|---|---|---|---|---|---|---|
| 1 |  Llama Nemotron Nano VL 8B Average Score ↑ | Llama | 60.20 | Open source | 无日期 来源 | |
| 2 | InternVL-3-14B Average Score ↑ | InternVL | 52.60 | Open source | 无日期 来源 | |
| 3 |  Gemini-Pro Average Score ↑ | Gemini | 51.90 | Closed / proprietary | 无日期 来源 | |
| 4 | Qwen2-VL-7B Average Score ↑ | Qwen2 | 51.40 | Open source | 无日期 来源 | |
| 5 | InternVL2.5-26B Average Score ↑ | InternVL2.5 | 49.40 | Open source | 无日期 来源 | |
| 6 | InternVL-3-8B Average Score ↑ | InternVL | 49 | Open source | 无日期 来源 | |
| 7 |  GPT-4V Average Score ↑ | GPT | 47.70 | Closed / proprietary | 无日期 来源 | |
| 8 | Ovis2-8B Average Score ↑ | Ovis2 | 47.70 | Open source | 无日期 来源 | |
| 9 | InternVL2-26B Average Score ↑ | InternVL2 | 46.80 | Open source | 无日期 来源 | |
| 10 | Step-1V Average Score ↑ | Step | 46.70 | Closed / proprietary | 无日期 来源 | |
| 11 | Qwen2.5-VL-7B Average Score ↑ | Qwen2.5 | 46.70 | Open source | 无日期 来源 | |
| 12 | GPT-4o Average Score ↑ | GPT | 46.50 | Closed / proprietary | 无日期 来源 | |
| 13 |  Claude3.5-sonnet Average Score ↑ | Claude3.5 | 45.20 | Closed / proprietary | 无日期 来源 | |
| 14 | MiniCPM-o-2.6 Average Score ↑ | MiniCPM | 45.10 | Open source | 无日期 来源 | |
| 15 | InternVL2.5-8B Average Score ↑ | InternVL2.5 | 45 | Open source | 无日期 来源 | |
| 16 | SAIL-VL-1.6-8B Average Score ↑ | SAIL | 44.80 | Open source | 无日期 来源 | |
| 17 | Deepseek-VL2-Small Average Score ↑ | Deepseek | 43.30 | Open source | 无日期 来源 | |
| 18 | GPT-4o-mini Average Score ↑ | GPT | 43 | Closed / proprietary | 无日期 来源 | |
| 19 | GLM-4V-Plus Average Score ↑ | GLM | 42.90 | Closed / proprietary | 无日期 来源 | |
| 20 | GLM-4V-9B Average Score ↑ | GLM | 42.60 | Open source | 无日期 来源 | |
| 21 | Kimi-VL-A3-16B Average Score ↑ | Kimi | 41.10 | Open source | 无日期 来源 | |
| 22 | Ovis1.6-3B Average Score ↑ | Ovis1.6 | 40.80 | Open source | 无日期 来源 | |
| 23 | MiniCPM-V-2.6 Average Score ↑ | MiniCPM | 40.30 | Open source | 无日期 来源 | |
| 24 | Pixtral-12B Average Score ↑ | Pixtral | 40.30 | Open source | 无日期 来源 | |
| 25 | InternVL2-8B Average Score ↑ | InternVL2 | 39.60 | Open source | 无日期 来源 | |
| 26 | LLaVA-OV-7B Average Score ↑ | LLaVA | 36.40 | Open source | 无日期 来源 | |
| 27 | Phi-4-MultiModal Average Score ↑ | Phi | 35.20 | Open source | 无日期 来源 | |
| 28 | Cambrian-1-8B Average Score ↑ | Cambrian | 34.70 | Open source | 无日期 来源 | |
| 29 | Molmo-7B Average Score ↑ | Molmo | 34.50 | Open source | 无日期 来源 | |
| 30 | Idefics3-8B Average Score ↑ | Idefics3 | 32.20 | Open source | 无日期 来源 | |
| 31 | LLaVA-Next-8B Average Score ↑ | LLaVA | 31.50 | Open source | 无日期 来源 | |
| 32 | XComposer2-4KHD Average Score ↑ | XComposer2 | 27.90 | Open source | 无日期 来源 | |
| 33 | Eagle-X5-7B Average Score ↑ | Eagle | 27.50 | Open source | 无日期 来源 | |
| 34 | Deepseek-VL-7B Average Score ↑ | Deepseek | 27.20 | Open source | 无日期 来源 | |
| 35 | mPLUG-Owl3 Average Score ↑ | mPLUG | 25.10 | Open source | 无日期 来源 | |
| 36 | TextMonkey Average Score ↑ | TextMonkey | 23.90 | Open source | 无日期 来源 | |
| 37 | VILA1.5-8B Average Score ↑ | VILA1.5 | 23.60 | Open source | 无日期 来源 | |
| 38 | Qwen-VL-chat Average Score ↑ | Qwen | 23.40 | Open source | 无日期 来源 | |
| 39 | Qwen-VL Average Score ↑ | Qwen | 23.30 | Open source | 无日期 来源 | |
| 40 | Monkey Average Score ↑ | Monkey | 23.10 | Open source | 无日期 来源 | |
| 41 | CogVLM-chat Average Score ↑ | CogVLM | 21.80 | Open source | 无日期 来源 | |
| 42 | DocOwl2 Average Score ↑ | DocOwl2 | 19.40 | Open source | 无日期 来源 | |
| 43 | EMU2-chat Average Score ↑ | EMU2 | 18.80 | Open source | 无日期 来源 | |
| 44 | Janus-1.3B Average Score ↑ | Janus | 18.70 | Open source | 无日期 来源 | |
| 45 | Yi-VL-6B Average Score ↑ | Yi | 17.30 | Open source | 无日期 来源 | |
| 46 | TextHarmony Average Score ↑ | TextHarmony | 16 | Open source | 无日期 来源 | |
| 47 | LLaVAR Average Score ↑ | LLaVAR | 15.30 | Open source | 无日期 来源 | |
| 48 | UReader Average Score ↑ | UReader | 13.70 | Open source | 无日期 来源 |