Chartographer: leaderboard
Chart question-answering (QA) benchmarks aim to pose questions that require visual reasoning to correctly answer, but models can often reach solutions through shortcuts or prior familiarity with a chart based on their own background knowledge. To strictly evaluate visual reasoning, we propose counterfactual charts where the chart-question task remains fixed, but underlying chart and the correspond
Metric: ChartQA OA (self-reported). Source: benchmarklist.com. Saturation forecast: Around March 2027. 15 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro | 94.7 | #152 |
| 2 | GPT-5.4 Mini | 94 | #212 |
| 3 | Gemini 2.5 Flash | 93.3 | #245 |
| 4 | Claude Sonnet 4.6 | 92.7 | #90 |
| 5 | GPT-5.4 | 92.7 | #79 |
| 6 | GPT-4.1 | 92.7 | #248 |
| 7 | Claude Haiku 4.5 | 90.7 | #280 |
| 8 | Qwen 3 VL 8B Instruct | 90.7 | #420 |
| 9 | GPT-4.1 Mini | 88 | #355 |
| 10 | GPT-4o | 86 | #343 |
| 11 | gemma-4-E4B-it | 86 | #566 |
| 12 | InternVL3-8B | 84.7 | #620 |
| 13 | Qwen 2.5 VL 7B Instruct | 84 | #654 |
| 14 | Pixtral-12B | 82.7 | #808 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=chartographer · How It Works · Data refreshed daily, snapshot 2026-10-20.