[Feature Request]: run ollama on localhost, don’t run in Docker.

[zhuang-weiming]

Is there an existing issue for the same feature request?

• I have checked the existing issues.

Is your feature request related to a problem?

Describe the feature you'd like

I love this project very much, Dify is a US AI platform, ragflow would be the best China AI platform I believe.
this is a convince feature, you know, every developer or user run ollama directly on localhost, like local Macbook or other kinds of laptop.
As the document of ragflow now - https://github.com/infiniflow/ragflow/blob/main/docs/guides/deploy_local_llm.mdx , it guide users to run ollama in Docker, could you consider to run the ollama directly on localhost computer.

Describe implementation you've considered

[KevinHuSh]

No matter where Ollama run, as long as the network is accessible, RAGFlow can utilize it anyway.

[zhuang-weiming]

No matter where Ollama run, as long as the network is accessible, RAGFlow can utilize it anyway.

Good to know, when I setup the LLM via ollama on localhost local machine, like the configue is: embedding / deepseek-r1:14b / http://host.docker.internal:11434, an error popup as below on the page of https://demo.ragflow.io/user-setting/model :

**hint : 102
Fail to access embedding model(deepseek-r1:14b).[Errno -2] Name or service not known**

could you kindly help?

[liaoweiguo]

No matter where Ollama run, as long as the network is accessible, RAGFlow can utilize it anyway.

Good to know, when I setup the LLM via ollama on localhost local machine, like the configue is: embedding / deepseek-r1:14b / http://host.docker.internal:11434, an error popup as below on the page of https://demo.ragflow.io/user-setting/model :

**hint : 102
Fail to access embedding model(deepseek-r1:14b).[Errno -2] Name or service not known**

could you kindly help?

这个不是嵌入模型

[KevinHuSh]

demo.ragflow.io and host.docker.internal:11434 are not pingable.

[zhuang-weiming]

got you, will install ragflow in my local env.
Please help to close this issue.
Thank you~