Announced on June 12, 2026, the PP-OCRv6 series includes Tiny, Small, and Medium models ranging from 1.5 million to 34.5 million parameters. PaddlePaddle reports support for 50 languages and CPU inference up to 5.2 times faster with OpenVINO.
PaddleOCR announced the PP-OCRv6 series on June 12, 2026. It includes Tiny, Small, and Medium models ranging from 1.5 million to 34.5 million parameters; PaddlePaddle reports support for 50 languages and new OCR scenarios. The company also reports CPU inference up to 5.2 times faster with OpenVINO.
The announcement highlights useful dimensions for evaluating OCR deployments: model size, language coverage, and inference options. To verify the results, consult the original announcement and documentation, checking which models and hardware and test conditions underpin the speed comparison; the published text does not specify those conditions. If using AI to study or apply the material, avoid sending identifiable documents until you have checked your organization’s data policy.