Xinference Deploy Model (Dify & RagFlow) - digmouse’s Ai journey

Install

pip install "xinference[all]"

Run

xinference -H 127.0.0.1 -p 9997
or
xinference-local --host 127.0.0.1 --port 9997 --endpoint-url http://127.0.0.1:9997

Select Model

Model Api


#RagFlow
http://localhost:9997/v1/rerank
or
http://host.docker.internal:9997/v1/rerank

#Dify
http://localhost:9997
or
http://host.docker.internal:9997

文章评论