PaddlePaddle/PaddleOCR

View on GitHub

Turn any PDF or image document into structured data for your AI. A powerful, lightweight OCR toolkit that bridges the gap between images/PDFs and LLMs. Supports 100+ languages.

Stars: 89.8k
Forks: 11.4k
Open beginner issues: 0
Indexed issues: 168
Avg merge: 1d 15h
Merged PRs (30d): 2

Dominant language: Python
License: Apache-2.0
Last GitHub push: Sep 16, 2026
Latest indexed: Sep 19, 2026
Contributing guide: No contributing guide
Code of conduct: No code of conduct
Beginner labels: No beginner labels indexed

Beginner friendly Everything

Open All

168 open issues indexed

Text on stderr, but not in markdown output. Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

PaddlePaddle/PaddleOCR#18371 · 1 comment · Opened Sep 18, 2026
Error in "concatenate_markdown_pages" example in Usage tutorial for PPStructureV3: TypeError: write() argument must be str, not MarkdownResult Open

Difficulty 2/5 1-3 hours Newbie friendliness 78/100

PaddlePaddle/PaddleOCR#18369 · Opened Sep 17, 2026
Broken documentation link in langchain-paddleocr/README.md Open

Difficulty 1/5 Under an hour Newbie friendliness 90/100

PaddlePaddle/PaddleOCR#18367 · Opened Sep 16, 2026
PP-OCRv6 训练可以使用图片高度为32的？对识别影响大？ Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

PaddlePaddle/PaddleOCR#18366 · Opened Sep 16, 2026
PP-OCRv6/v5 unified rec ONNX 模型非 ASCII 输出为 CP1252 渲染的 UTF-8 字节（PP-OCRv6_small_rec_onnx） Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

PaddlePaddle/PaddleOCR#18364 · Opened Sep 16, 2026
resize_norm_img 直接resize问题，宽高比大于10的数据丢弃代码怎么没找到呢 Open

Difficulty 3/5 1-2 days Newbie friendliness 45/100

PaddlePaddle/PaddleOCR#18358 · 2 comments · Opened Sep 14, 2026
PIR optimization does not yet support the Kunlun P800 server environment. Open

Difficulty 4/5 3-5 days Newbie friendliness 45/100

PaddlePaddle/PaddleOCR#18353 · Opened Sep 9, 2026
arabic_PP-OCRv5_mobile_rec: native aspect-ratio width under-allocates CTC steps for full-width lines (36.7% → 12.9% CER) Open

Difficulty 4/5 3-5 days Newbie friendliness 52/100

PaddlePaddle/PaddleOCR#18349 · 3 comments · Opened Sep 7, 2026
识别不准确 Open

Difficulty 4/5 3-5 days Newbie friendliness 45/100

PaddlePaddle/PaddleOCR#18347 · 4 comments · Opened Sep 5, 2026
ppocrv6 ">=" not supported between instance of 'list' and float Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

PaddlePaddle/PaddleOCR#18343 · Opened Sep 3, 2026
容器内推理时 GPU 推理报错 MMU Fault: ENGINE GRAPHICS GPC2 GPCCLIENT_T1_0， fallback 到 CPU，重启容器解决 Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

PaddlePaddle/PaddleOCR#18328 · Opened Aug 26, 2026
[PP-OCRv6][Performance] High CPU inference latency and low CPU utilization - need Small model accuracy with lower latency Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

PaddlePaddle/PaddleOCR#18320 · 1 comment · Opened Aug 20, 2026
PaddleOCR's recognition performance on the following documents is still not satisfactory Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

PaddlePaddle/PaddleOCR#18318 · Opened Aug 20, 2026
npu 容器启动失败 Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

PaddlePaddle/PaddleOCR#18315 · Opened Aug 18, 2026
PaddleOcr-Vl-1.6 智能版面分析识别表格时，内容缺失 Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

PaddlePaddle/PaddleOCR#18314 · Opened Aug 18, 2026
字符识别错误或者无法识别 Open

Difficulty 4/5 3-5 days Newbie friendliness 45/100

PaddlePaddle/PaddleOCR#18313 · 1 comment · Opened Aug 18, 2026
`save_to_markdown` overwrites images with same coordinates on different pages Open

Difficulty 3/5 1-2 days Newbie friendliness 58/100

PaddlePaddle/PaddleOCR#18305 · 1 comment · Opened Aug 12, 2026
AMD ROCm paddleocr-vl:paddleocr3.6-amd-gpu-offline fails to start Open

PaddlePaddle/PaddleOCR#18300 · 1 assignee · Opened Aug 4, 2026
PP-OCRv6识别准确率问题 Open

PaddlePaddle/PaddleOCR#18299 · 2 comments · 1 assignee · Opened Aug 3, 2026
Verify evals on Papers with Code Open

PaddlePaddle/PaddleOCR#18298 · 1 assignee · Opened Jul 30, 2026
PP-OCRV6竖排文本不识别标点符号，横排是没有问题的 Open

PaddlePaddle/PaddleOCR#18297 · 1 assignee · Opened Jul 30, 2026
在A10GPU上训练PP-DocLayoutV3模型，使用官方数据集完训练完成后，在基于PaddleOCR-VL 的服务化部署（PaddleOCR/deploy/paddleocr_vl_docker/hps）替换PP-DocLayoutV3模型，启动paddleocr-vl-pipeline服务后报以下错误，求助分析原因 Open

PaddlePaddle/PaddleOCR#18296 · 1 assignee · Opened Jul 30, 2026
PP-OCRv6 models are not registered in PaddleX when using PP-StructureV3 pipeline Open

PaddlePaddle/PaddleOCR#18295 · 1 comment · 1 assignee · Opened Jul 30, 2026
[SLANet_plus][ONNX] Please verify export path: Loop-body inputs shadow visible outer-scope values Open

PaddlePaddle/PaddleOCR#18286 · 1 assignee · Opened Jul 27, 2026
Android demo : Use OpenCV version 4.5.3.0 not version 4.5.3 Open

PaddlePaddle/PaddleOCR#18285 · 1 assignee · Opened Jul 26, 2026
[BUG] CUDA error(700) / cudaErrorIllegalAddress when running PPStructureV3 / PaddleOCR on Python 3.13 & CUDA 13 Driver Open

PaddlePaddle/PaddleOCR#18284 · 1 assignee · Opened Jul 26, 2026
paddlepaddle-gpu issue in H100 server Open

PaddlePaddle/PaddleOCR#18280 · 2 comments · 1 assignee · Opened Jul 23, 2026
[third-party] Add light-ocr to awesome projects Open

PaddlePaddle/PaddleOCR#18276 · 3 comments · 1 assignee · Opened Jul 22, 2026
Possible thread-safety issue: Concurrent PaddleOCR.predict() calls on a shared instance produce corrupted OCR results Open

PaddlePaddle/PaddleOCR#18268 · 1 comment · 1 assignee · Opened Jul 21, 2026
带底纹的征信图片会出现大量的0 Open

PaddlePaddle/PaddleOCR#18259 · 1 assignee · Opened Jul 15, 2026
【PaddleOCR-VL-1.6】OmniDocBench v1.6 单页在 llama.cpp/GGUF VLM 路径稳定 500：drop-cap 窄栏文本触发 peg-native 解析失败 Open

PaddlePaddle/PaddleOCR#18248 · 3 comments · 2 assignees · Opened Jul 7, 2026
PP-OCRv5 mobile det+rec silently drops most text in table rows separated by grid/border lines Open

PaddlePaddle/PaddleOCR#18244 · 2 comments · 2 assignees · Opened Jul 7, 2026
PP-DocLayoutV3基本上全部忽略参考文献 Open

PaddlePaddle/PaddleOCR#18241 · 2 comments · 2 assignees · Opened Jul 6, 2026
3.7.0版本训练的PPOCRV6的medium_det模型导出为推理模型后再转成onnx模型时候报错 Open

PaddlePaddle/PaddleOCR#18237 · 2 comments · 1 reaction · 2 assignees · Opened Jul 3, 2026
识别几次图片后报错，使用的最新的paddle3.3.1 paddleocrv6 显卡 RTX5060Ti Open

PaddlePaddle/PaddleOCR#18235 · 1 comment · 2 assignees · Opened Jul 1, 2026
图文混排的多页PDF文档只识别了第一页 Open

PaddlePaddle/PaddleOCR#18233 · 2 comments · 2 assignees · Opened Jul 1, 2026
PP-DocLayoutV3 版面识别问题 Open

PaddlePaddle/PaddleOCR#18215 · 6 comments · 1 assignee · Opened Jun 25, 2026
PaddleOCR-v6 文字方向检测失败 Open

PaddlePaddle/PaddleOCR#18214 · 2 comments · 1 assignee · Opened Jun 25, 2026
cpp_infer3.7 本地部署 PP-OCRv6 Open

PaddlePaddle/PaddleOCR#18213 · 7 comments · 1 assignee · Opened Jun 24, 2026
需要支持hps:paddlex3.7-gpu 镜像和部署 PP-OCRv6_medium模型加载问题 Open

PaddlePaddle/PaddleOCR#18212 · 1 assignee · Opened Jun 24, 2026
飞桨OCR与torch环境不兼容 Open

PaddlePaddle/PaddleOCR#18208 · 1 comment · 1 assignee · Opened Jun 24, 2026
Segmentation fault Open

PaddlePaddle/PaddleOCR#18204 · 1 comment · 1 assignee · Opened Jun 23, 2026
paddlex 使用cpu cv 预处理偶发报错 Open

PaddlePaddle/PaddleOCR#18202 · 1 assignee · Opened Jun 23, 2026
paddleocr With lang="en" I still get Chinese characters Open

PaddlePaddle/PaddleOCR#18201 · 1 comment · 1 assignee · Opened Jun 23, 2026
PP-OCRv5: use_textline_orientation=True corrupts bold/large text lines on GPU (CPU unaffected, resolution-dependent) Open

PaddlePaddle/PaddleOCR#18195 · 1 assignee · Opened Jun 22, 2026
PaddleOCR-VL HPS: returnMarkdownImages=false is ineffective with default PaddleX 3.6 SDK Open

PaddlePaddle/PaddleOCR#18194 · 1 assignee · Opened Jun 22, 2026
PP-OCRv6_medium_rec 转onnx [ERROR][Paddle2ONNX] Due to the unsupported operators, the conversion is aborted. Open

PaddlePaddle/PaddleOCR#18190 · 7 comments · 1 assignee · Opened Jun 22, 2026
cpp_infer Crashes When Using PP-OCRv6 Open

PaddlePaddle/PaddleOCR#18178 · 7 comments · 1 assignee · Opened Jun 17, 2026
证书到期 Open

PaddlePaddle/PaddleOCR#18177 · 3 comments · 1 assignee · Opened Jun 17, 2026
PaddleOCR seems to require root? Open

stale

PaddlePaddle/PaddleOCR#18175 · 3 comments · 1 assignee · Opened Jun 17, 2026

PaddlePaddle/PaddleOCR

Get new issues in your inbox