Skip to content

Benchmark

Real5 Omnidocbench

Scores are percentages. Higher is better.

Models measured
9
Scores recorded
9
Sources
1
IQ weight
1.53

Score distribution

83%
Mean
88.9%
Median
10.1%
Std dev
73%
10th percentile
90.3%
90th percentile
61.8%
Min
61.8%93.2%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.53 for chat models, measured over 5 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1PaddleOCR-VL-1.6· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim93.2%
2GLM OCR· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim90.3%
3Kimi K2.6· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim89.8%
4Kimi K2.5· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim89.1%
5Qwen3 VL 235B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim88.9%
6Qwen2.5 VL 72B Instruct· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim86.9%
7DeepSeek-OCR· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim74%
8DeepSeek OCR 2· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim73%
9dolphin· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim61.8%

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