ISO 15924 · Hani · logographic
Chinese/CJK
The Chinese/CJK writing system (ISO 15924 Hani), a logographic: 31 languages use it, 103,351 Unicode characters. Best OCR accuracy: 86% Acc@5 (dots.mocr).
Languages written in Chinese/CJK
- Bouyeipcc
- Central Baibca
- Central Hongshuihe Zhuangzch
- Classical Chineselzh
- Dai Zhuangzhd
- Eastern Hongshuihe Zhuangzeh
- Guibei Zhuangzgb
- Guibian Zhuangzgn
- Hakka Chinesehak
- Honihow
- Japanesejpn
- Koreankor
- Lianshan Zhuangzln
- Liujiang Zhuangzlj
- Liuqian Zhuangzlq
- Mandarin Chinesecmn
- Min Nan Chinesenan
- Minz Zhuangzgm
- Nong Zhuangzhn
- Qiubei Zhuangzqe
- Southern Baibfs
- Suiswi
- Wu Chinesewuu
- Yang Zhuangzyg
- Yongnan Zhuangzyn
- Youjiang Zhuangzyj
- Yue Chineseyue
- Zuojiang Zhuangzzj
- Chinesezho
OCR results
- dots.mocr: Acc@5 86.5%, CER 2.8
- Gemini 3.1 Flash-Lite: Acc@5 85.5%, CER 2.5
- HunyuanOCR: Acc@5 83.0%, CER 3.5
- GLM-OCR: Acc@5 82.0%, CER 2.7
- dots.ocr: Acc@5 81.5%, CER 3.5
- olmOCR-2: Acc@5 78.2%, CER 3.2
- Qwen3-VL-8B: Acc@5 78.0%, CER 3.5
- RolmOCR: Acc@5 68.8%, CER 5.2
- FireRed-OCR: Acc@5 66.8%, CER 6.3
- PaddleOCR-VL-1.5: Acc@5 60.8%, CER 7.2
- GPT-4.1: Acc@5 55.0%, CER 7.5
- DeepSeek-OCR-2: Acc@5 48.5%, CER 9.2
- Nanonets-OCR2-3B: Acc@5 46.5%, CER 10.8
- LightOnOCR-2-1B: Acc@5 26.2%, CER 17.4