LLM Stats (CC-OCR): leaderboard
Comprehensive OCR benchmark testing text recognition in complex real-world scenes including curved text, artistic fonts, and multilingual content.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Estimated already saturated. 18 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.6 Plus | 83.4 | #120 |
| 2 | Qwen 3 VL 235B A22B Instruct | 82.2 | #273 |
| 3 | Qwen 3.6 35B A3B | 81.9 | #226 |
| 4 | Qwen 3.5 122B A10B | 81.8 | #173 |
| 5 | Qwen 3 VL 235B A22B (Thinking) | 81.5 | #238 (Qwen 3 VL 235B A22B) |
| 6 | Qwen 3.6 27B | 81.2 | #170 |
| 7 | Qwen 3.5 27B | 81 | #187 |
| 8 | Qwen 3.5 35B A3B | 80.7 | #256 |
| 9 | Qwen 3 VL 30B A3B Instruct | 80.7 | #384 |
| 10 | Qwen 3 VL 32B Instruct | 80.3 | #283 |
| 11 | Qwen 3 VL 8B Instruct | 79.9 | #420 |
| 12 | Qwen 2.5 VL 72B Instruct | 79.8 | #377 |
| 13 | Qwen 2.5 VL 7B Instruct | 77.8 | #654 |
| 14 | Qwen 3 VL 30B A3B (Thinking) | 77.8 | #351 (Qwen 3 VL 30B A3B) |
| 15 | Qwen 2.5 VL 32B Instruct | 77.1 | #453 |
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-cc-ocr · How It Works · Data refreshed daily, snapshot 2026-10-20.