ChartArena - Flowchart (Chinese): leaderboard
Metric: mAP-high (%) on Chinese flowcharts, each model parses the chart image into its native structured format, normalized format-agnostically to (header, entity, value) triples for numeric charts or to labeled graphs and trees for flowcharts and mind maps; mean average precision over similarity thresholds at the high tolerance level, averaged over three visual styles (digital renderings, printed photos, hand-drawn photos); higher is better. Source: arxiv.org. Saturation forecast: Around December 2026. 38 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 63.1 |
| 2 | Seed 2.0 Pro (Non-reasoning) | 61.3 |
| 3 | Kimi K2.5 (Non-reasoning) | 57.9 |
| 4 | Gemini 2.5 Pro | 57.8 |
| 5 | Qwen 3.5 35B A3B (Thinking) | 56.5 |
| 6 | MiMo-V2-Omni | 51 |
| 7 | Qwen 3 VL 235B A22B Instruct | 49.8 |
| 8 | Gemma 4 31B (IT) | 44 |
| 9 | Qwen 2.5 VL 72B Instruct | 43.6 |
| 10 | Qwen 3 VL 8B Instruct | 41.5 |
| 11 | Qwen 2.5 VL 32B Instruct | 40.1 |
| 12 | GLM-4.5V | 39.6 |
| 13 | Qwen 3.5 9B | 38.1 |
| 14 | Qwen 3 VL 8B (Thinking) | 38.1 |
| 15 | Qwen 3 VL 4B Instruct | 35.9 |
Interactive version: theaggregate.ai/benchmark?slug=chartarena-flowchart-chinese · How It Works · Data refreshed daily, snapshot 2026-09-29.