Benchmark
ChartQA
Vision benchmarks measure how well models understand images. Scores are percentages. Higher is better.
- Models measured
- 32
- Scores recorded
- 34
- Sources
- 3
- IQ weight
- 0.80
Score distribution
- 69.4%
- Mean
- 83%
- Median
- 26.2%
- Std dev
- 30.3%
- 10th percentile
- 88.8%
- 90th percentile
- 0%
- Min
This benchmark has a weight of 0.80 for chat models, measured over 31 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Claude 3.5 Sonnet· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 90.8% |
| 2 | Llama 4 Maverick· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 90% |
| 3 | Qwen2.5 VL 72B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 89.5% |
| 4 | Nova Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 89.2% |
| 5 | Llama 4 Scout· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 88.8% |
| 6 | Llama 4 Scoutinstruction-tuned table· Reported by Meta· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 88.8% |
| 7 | Qwen2-VL-72B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 88.3% |
| 8 | Pixtral Large· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 88.1% |
| 9 | Mistral Small 3.2 24B Instruct2506· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 87.4% |
| 10 | Qwen2.5-VL 7B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 87.3% |
| 11 | Nova Lite· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 86.8% |
| 12 | DeepSeek VL2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 86% |
| 13 | internvl3-8b· OpenVLM Leaderboard data | 85.8% |
| 14 | GPT 4o· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 85.7% |
| 15 | Qwen2.5-Omni 7B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 85.3% |
| 16 | InternVL2 26B· OpenVLM Leaderboard data | 85.1% |
| 17 | Llama 3.2 11B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 83.4% |
| 18 | Phi 4 Multimodal Instruct· OpenVLM Leaderboard dataValues from 2 sources
| 83% |
| 19 | MiniCPM-V 2.6· OpenVLM Leaderboard data | 82.4% |
| 20 | Phi-3.5 Vision Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 81.8% |
| 21 | Pixtral 12B2409· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 81.8% |
| 22 | InternVL2 4B· OpenVLM Leaderboard data | 80.7% |
| 23 | Gemma 3 27B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 78% |
| 24 | Gemma 3 12B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 75.7% |
| 25 | Pixtral 12B· OpenVLM Leaderboard data | 72.2% |
| 26 | Gemma 3 4B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 68.8% |
| 27 | Molmo 7B-D· OpenVLM Leaderboard data | 43.7% |
| 28 | Gemma 3 4B· OpenVLM Leaderboard data | 33.7% |
| 29 | GPT-4.1 mini· OpenVLM Leaderboard data | 33.7% |
| 30 | GPT 4.1· OpenVLM Leaderboard data | 33.3% |
| 31 | GPT-4.1 nano· OpenVLM Leaderboard data | 30.3% |
| 32 | Gemini 1.5 Flash· OpenVLM Leaderboard data | 29.7% |
| 33 | kosmos-2· OpenVLM Leaderboard data | 5.9% |
| 34 | Llama 3.2 11B Vision Instruct· OpenVLM Leaderboard data | 0% |
Data credits: Data by LLM Stats · OpenVLM Leaderboard (OpenCompass / VLMEvalKit), Apache-2.0 (licence) · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.