Benchmark
CharXiv-D
Vision benchmarks measure how well models understand images. Scores are percentages.
All vision benchmarksScores from LLM Stats data.
- Models measured
- 14
- Scores recorded
- 15
- Sources
- 1
- IQ weight
- 2.27
Score distribution
- 87.8%
- Mean
- 88%
- Median
- 5%
- Std dev
- 83%
- 10th percentile
- 94.6%
- 90th percentile
- 73.9%
- Min
This benchmark has a weight of 2.27 for chat models, measured over 14 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 | Doubao Seed 2.1 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 95.5% |
| 2 | Doubao Seed 2.1 Turbo· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 94.6% |
| 3 | Seed 2.0 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 91.9% |
| 4 | Qwen3 VL 32B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 90.5% |
| 5 | Qwen3 VL 32B Thinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 90.2% |
| 6 | GPT-4.5 Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 90% |
| 7 | GPT-4.1 mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 88.4% |
| 8 | Command A Plus 05 2026· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 88% |
| 9 | GPT 4.1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 87.9% |
| 10 | qwen3-vl-30bthinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 86.9% |
| 11 | Qwen: Qwen3 VL 8B Thinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 85.9% |
| 12 | qwen3-vl-30b· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 85.5% |
| 13 | GPT 4o· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 85.3% |
| 14 | Qwen3 VL 8B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 83% |
| 15 | GPT-4.1 nano· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 73.9% |
Data credits: Data by LLM Stats