ParseBench - Charts: leaderboard
LlamaIndex's ParseBench scores document parsing of human-verified enterprise pages from insurance, finance and government documents with rule-based checks; VLM rows only, parser products excluded. This board reports the Charts dimension: exact data-point extraction with correct series and axis labels from bar, line, pie and compound charts.
Metric: ChartDataPointMatch Score. Source: parsebench.ai. Saturation forecast: Around February 2028. 121 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5.5 (High) | 70.89 | #3 (Claude Opus 5.5) |
| 2 | GPT-5.6 Sol (High) | 68.68 | #22 (GPT-5.6 Sol) |
| 3 | GPT-5.6 Terra (Medium) | 67.86 | #43 (GPT-5.6 Terra) |
| 4 | GPT-5.6 Terra (Low) | 67.71 | #43 (GPT-5.6 Terra) |
| 5 | Claude Fable 5.1 | 66.93 | #10 |
| 6 | GPT-5.4 (Non-reasoning) | 65.57 | #85 (GPT-5.4) |
| 7 | GPT-5.5 (Medium) | 64.84 | #34 (GPT-5.5) |
| 8 | Claude Opus 5.5 | 64.12 | #3 |
| 9 | Gemini 3 Flash (High) | 64.05 | #105 (Gemini 3 Flash) |
| 10 | Gemini 3 Flash (Minimal) | 63.63 | #105 (Gemini 3 Flash) |
| 11 | GPT-5.6 Terra (Non-reasoning) | 61.68 | #43 (GPT-5.6 Terra) |
| 12 | Gemini 3 Flash (Medium) | 61.56 | #105 (Gemini 3 Flash) |
| 13 | Claude Sonnet 5 | 60.53 | #67 |
| 14 | GPT-6.1 Terra (Medium) | 59.71 | #14 (GPT-6.1 Terra) |
| 15 | GPT-5.5 (Non-reasoning) | 58.8 | #34 (GPT-5.5) |
No result here: #9 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=parsebench-charts · How It Works · Data refreshed daily, snapshot 2026-11-07.