OCR models / Baidu (PaddlePaddle)
PaddleOCR-VL 1.5
Baidu (PaddlePaddle) · released 2026-01-29 · data checked 2026-11-06
PaddleOCR-VL 1.5 is a 960M-parameter OCR and document-parsing model from Baidu (PaddlePaddle) (CN), released 2026-01-29 under the Apache-2.0 license. It scores 94.93 on OmniDocBench v1.6 (#10 of 33). It needs about 2.9 GB of memory at 16-bit.
Previous release of PaddleOCR-VL, succeeded by 1.6. Pairs the 0.9B model with the PP-DocLayout layout detector.
OmniDocBench v1.6
94.93
#10 of 33 models
olmOCR-Bench
—
Not reported
Parameters
960M
OCR specialist
Memory to run
~2.9 GB
16-bit weights plus about 1 GB
Specifications
- Lab
- Baidu (PaddlePaddle) (CN)
- Type
- OCR / document parsing specialist
- Total parameters
- 960M
- License
- Apache-2.0
- Memory to run
- ~2.9 GB at 16-bit
- Released
- 2026-01-29
Closest alternatives
- HunyuanOCR 1.5 (1.1B, 94.74 on OmniDocBench, Tencent Hunyuan Community License)
- GLM-OCR (1.3B, 95.22 on OmniDocBench, MIT)
- MiniCPM-V 4.7 35B-A3B (35.2B, 94.24 on OmniDocBench, Apache-2.0)
- MinerU2.5 Pro (2605) (1.2B, 95.72 on OmniDocBench, Apache-2.0)
Sources
Something wrong here? Report it.