‹ PublicAI Index
The LLM benchmark aggregator.
PaddleOCR VL
Other
Strongest in Document parsing (#64 of 76), weakest in Knowledge (#175 of 190). Above par in 0 of 2 scopes. Among the models it meets almost everywhere, it finishes behind Claude Opus 5.5 and Claude Fable 5.1 and ahead of DeepSeek OCR 2 and Gemma 4 E4B.
§ 1 · Profile
What it is good at
Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.
Knowledge40.1−22.7#175/1901/3
Document parsing39.4−24.3#64/761/1
§ 2 · Head to head
What it beats, and what beats it
The same models turn up scope after scope. Counted once: where both were placed, who finished higher. Bars run right for PaddleOCR VL, left for the other.
§ 3 · Sources
Where the numbers come from
1 publication, 6 figures. Every one links to the page it was read from.
ParseBench · Overall 40.9%ParseBench · Tables 67.1%ParseBench · Charts 0.9%ParseBench · Content faithfulness 82.7%ParseBench · Semantic formatting 54%ParseBench · Visual grounding 0%
Badge
[](https://publicai.io/model-index/m/paddleocr-vl)