LLM Stats (OCRBench_V2): leaderboard
LLM Stats aggregate for OCRBench v2, testing OCR and document understanding on harder multilingual text, table, formula, handwriting, and scene-text recognition tasks.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Around 2029. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.7 Plus | 67.1 | #92 |
| 2 | Seed 2.1 Pro | 63.2 | #67 |
| 3 | Seed 2.1 Turbo | 62.8 | #86 |
| 4 | Seed 2.0 Mini | 58.5 | #276 |
| 5 | Qwen2.5-Omni-7B | 57.8 | #624 |
| 6 | Nova 2.0 Lite | 56.1 | #417 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-ocrbench-v2 · How It Works · Data refreshed daily, snapshot 2026-10-20.