myhloli dba28389bc refactor(model): update model downloads and disable unused models 8 months ago
..
Dockerfile c734f4debe refactor(web_api): Optimize `Dockerfile` 9 months ago
README.md d9406e5bd6 docs(web_api): Update `README.md` 9 months ago
app.py 2bdb544553 update api path and documents 8 months ago
download_models.py dba28389bc refactor(model): update model downloads and disable unused models 8 months ago
entrypoint.sh c734f4debe refactor(web_api): Optimize `Dockerfile` 9 months ago
magic-pdf.json c734f4debe refactor(web_api): Optimize `Dockerfile` 9 months ago
requirements.txt c734f4debe refactor(web_api): Optimize `Dockerfile` 9 months ago

README.md

基于MinerU的PDF解析API

  • MinerU的GPU镜像构建
  • 基于FastAPI的PDF解析接口

构建方式

docker build -t mineru-api .

或者使用代理:

docker build --build-arg http_proxy=http://127.0.0.1:7890 --build-arg https_proxy=http://127.0.0.1:7890 -t mineru-api .

启动命令

docker run --rm -it --gpus=all -v ./paddleocr:/root/.paddleocr -p 8000:8000 mineru-api

初次调用 API 时会自动下载 paddleocr 的模型(约数十 MB),其余模型已包含在镜像中。

测试参数

访问地址:

http://localhost:8000/docs
http://127.0.0.1:8000/docs

旧版镜像地址

阿里云地址:docker pull registry.cn-beijing.aliyuncs.com/quincyqiang/mineru:0.1-models

dockerhub地址:docker pull quincyqiang/mineru:0.1-models

旧版截图

启动命令

具体截图请见博客:https://blog.csdn.net/yanqianglifei/article/details/141979684

启动日志

测试参数

解析效果