CC-OCR
South China University of Technology
Benchmark for OCR across multi-scene, multilingual, and document parsing tasks.
12
Total Results
5
Models Tested
4
Metrics
2025-12-21
Last Updated
Multi-Scene F1
F1 score on multi-scene text reading
Higher is better
| Rank | Model | Score | Source |
|---|---|---|---|
| 1 | gemini-15-pro Multi-Scene Text Reading - Overall F1 score | 83.25% | alphaxiv-leaderboard |
| 2 | qwen2-vl-72b | 77.95% | alphaxiv-leaderboard |
| 3 | internvl2-76b | 76.92% | alphaxiv-leaderboard |
| 4 | gpt-4o | 76.4% | alphaxiv-leaderboard |
| 5 | claude-35-sonnet | 72.87% | alphaxiv-leaderboard |
KIE F1
F1 score on key information extraction
Higher is better
| Rank | Model | Score | Source |
|---|---|---|---|
| 1 | qwen2-vl-72b Key Information Extraction - Overall F1 score | 71.76% | alphaxiv-leaderboard |
| 2 | gemini-15-pro | 67.28% | alphaxiv-leaderboard |
| 3 | claude-35-sonnet | 64.58% | alphaxiv-leaderboard |
| 4 | gpt-4o | 63.45% | alphaxiv-leaderboard |
Multilingual F1
F1 score on multilingual text (10 languages)
Higher is better
| Rank | Model | Score | Source |
|---|---|---|---|
| 1 | gemini-15-pro Multilingual Text Reading - 10 languages | 78.97% | alphaxiv-leaderboard |
| 2 | gpt-4o | 73.44% | alphaxiv-leaderboard |
Document Parsing
Average score on document parsing
Higher is better
| Rank | Model | Score | Source |
|---|---|---|---|
| 1 | gemini-15-pro Document Parsing - Average Score | 62.37 | alphaxiv-leaderboard |