Recent studyBlind TTS Elo is live. Compare two anonymous voice samples, vote after listening, and help separate real preference signal from noise.Vote in the study ->
Codesota · Benchmark · CC-OCRHome/Leaderboards/CC-OCR
South China University of Technology

CC-OCR.

Multi-scene text reading, key information extraction, multilingual text, and document parsing benchmark.

Paper ↗Leaderboard ↓
§ 01 · SOTA history

Year over year.

§ 02 · Leaderboard

Results by metric.

Found a wrong score or missing run?
Use row edits to send a sourced correction into moderation.
Add / edit result ↗Report issue ↗

Multi-Scene F1

Multi Scene F1 is the reported evaluation metric for CC-OCR. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for Multi-Scene F1verifiedpapervendorcommunityunverified
RankModelTrustScoreYearLinksEdit
01Gemini 1.5 Pro
Multi-Scene Text Reading - Overall F1 score
unverified83.252025Source ↗Edit result
02gemini-15-pro
Multi-Scene Text Reading - Overall F1 score
paper83.252025Source ↗Edit result
03qwen2-vl-72bpaper77.952025Source ↗Edit result
04Qwen2-VL 72Bunverified77.952025Source ↗Edit result
05InternVL2-76Bunverified76.922025Source ↗Edit result
06gpt-4opaper76.42025Source ↗Edit result
07Claude 3.5 Sonnetunverified72.872025Source ↗Edit result
08claude-35-sonnetpaper72.872025Source ↗Edit result
09GOT-OCR2.0verified612024Source ↗Edit result
10TextMonkeyverified56.882024Source ↗Edit result
11Florence-2-Largeverified49.242024Source ↗Edit result
12KOSMOS-2.5verified47.552024Source ↗Edit result

Multilingual F1

Multilingual F1 is the reported evaluation metric for CC-OCR. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for Multilingual F1verifiedpapervendorcommunityunverified
RankModelTrustScoreYearLinksEdit
01gemini-15-pro
Multilingual Text Reading - 10 languages
paper78.972025Source ↗Edit result
02Gemini 1.5 Pro
Multilingual Text Reading - 10 languages
unverified78.972025Source ↗Edit result
03gpt-4opaper73.442025Source ↗Edit result
04Qwen2-VL 72B
Multilingual Text Reading - 10 languages
verified71.142024Source ↗Edit result
05Claude 3.5 Sonnet
Multilingual Text Reading - 10 languages
verified65.682024Source ↗Edit result
06Florence-2-Large
Multilingual Text Reading - 10 languages
verified49.72024Source ↗Edit result
07InternVL2-76B
Multilingual Text Reading - 10 languages
verified46.572024Source ↗Edit result
08KOSMOS-2.5
Multilingual Text Reading - 10 languages
verified36.232024Source ↗Edit result
09GOT-OCR2.0
Multilingual Text Reading - 10 languages
verified24.952024Source ↗Edit result

KIE F1

Kie F1 is the reported evaluation metric for CC-OCR. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for KIE F1verifiedpapervendorcommunityunverified
RankModelTrustScoreYearLinksEdit
01qwen2-vl-72b
Key Information Extraction - Overall F1 score
paper71.762025Source ↗Edit result
02Qwen2-VL 72B
Key Information Extraction - Overall F1 score
unverified71.762025Source ↗Edit result
03gemini-15-propaper67.282025Source ↗Edit result
04Gemini 1.5 Prounverified67.282025Source ↗Edit result
05Claude 3.5 Sonnetunverified64.582025Source ↗Edit result
06claude-35-sonnetpaper64.582025Source ↗Edit result
07GPT-4ounverified63.452025Source ↗Edit result
08InternVL2-76B
Key Information Extraction - Overall F1 score
verified61.62024Source ↗Edit result

Document Parsing

Document Parsing is the reported evaluation metric for CC-OCR. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for Document Parsingverifiedpapervendorcommunityunverified
RankModelTrustScoreYearLinksEdit
01Gemini 1.5 Pro
Document Parsing - Average Score
unverified62.372025Source ↗Edit result
02gemini-15-pro
Document Parsing - Average Score
paper62.372025Source ↗Edit result
03Qwen2-VL 72B
Document Parsing - Average Score
verified53.782024Source ↗Edit result
04GPT-4o
Document Parsing - Average Score
verified53.32024Source ↗Edit result
05Claude 3.5 Sonnet
Document Parsing - Average Score
verified47.792024Source ↗Edit result
06GOT-OCR2.0
Document Parsing - Average Score
verified39.182024Source ↗Edit result
07InternVL2-76B
Document Parsing - Average Score
verified35.332024Source ↗Edit result
§ 04 · Submit a result

Add to the leaderboard.

← Back to Leaderboards