Skip to content

Benchmark

ParseBench

Scores are percentages. Higher is better.

Models measured
16
Scores recorded
16
Sources
2
IQ weight
0.84

Score distribution

56.8%
Mean
56.6%
Median
14.4%
Std dev
44.1%
10th percentile
75.1%
90th percentile
29.6%
Min
29.6%79.2%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.84 for chat models, measured over 6 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Parse V5.0· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim79.2%
2LightOnOCR-3 4B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim75.1%
3Lightonocr 3 0.8b· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim74.6%
4Qwen 3.8 27B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim70.8%
5GLM 5.3 Flash· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim70.8%
6PaddleOCR-VL-1.6· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim67.4%
7HunyuanOCR· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim63.9%
8DeepSeek V4.1 Flash· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim56.6%
9Nemotron 3 Nano Omni 30b A3b Reasoning· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim48.5%
10LightOnOCR-2-1B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim48%
11Qwen3 VL 8B Instruct· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim46.8%
12Qianfan-OCR· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim46.2%
13Jina OCR v1· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim45.9%
14Qwen3.6 35B A3B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim44.1%
15DeepSeek OCR 2· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim41.2%
16GLM OCR· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim29.6%

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