PaddleX/3.3/pipeline_deploy/serving

[giscus[bot]]

PaddleX/3.3/pipeline_deploy/serving

https://paddlepaddle.github.io/PaddleX/3.3/pipeline_deploy/serving.html

[gokamisama]

docker pull ccr-2vdh3abv-pub.cnc.bj.baidubce.com/paddlex/hps:paddlex3.3-cpu 无法拉取显示：
Error response from daemon: failed to resolve reference "ccr-2vdh3abv-pub.cnc.bj.baidubce.com/paddlex/hps:paddlex3.3-cpu": ccr-2vdh3abv-pub.cnc.bj.baidubce.com/paddlex/hps:paddlex3.3-cpu: not found

[lilong1988]

没有这个镜像：paddlex/hps:paddlex3.3-gpu
另外想问一下，如果通过docker部署PaddleOCR-VL产线，此镜像是否通过vllm或者sglang运行vl模型？

[gokamisama]

https://paddlepaddle.github.io/PaddleX/latest/pipeline_usage/tutorials/ocr_pipelines/PaddleOCR-VL.html#311-docker

专门有vllm推理paddleocr-vl的镜像