Skip to content

Benchmark

Parsebench 2

Scores are percentages. Higher is better.

Models measured
9
Scores recorded
9
Sources
1
IQ weight
No weight published

Score distribution

62.3%
Mean
67.1%
Median
13.4%
Std dev
42.3%
10th percentile
75.9%
90th percentile
42.3%
Min
42.3%77%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1LightOnOCR-3 4B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim77%
2Lightonocr 3 0.8b· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim75.9%
3Qwen 3.8 27B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim73.4%
4PaddleOCR-VL-1.6· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim70.3%
5HunyuanOCR· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim67.1%
6unlimited-ocr-2· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim64.4%
7Jina OCR v1· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim48.2%
8Qwen3.5 4B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim42.3%
9Gemma 4 E4B-it· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim42.3%

Data credits: Hugging Face official benchmark leaderboards; licence varies by benchmark