ChatOllama. how to use base_url

[lc345]

llm = ChatOllama(
model="gemma3:1b",
base_url="http://...:8001/api/chat",
) This is the model I started with ollama on another server. I tested it and it works, but ChatOllama reports an error: ollama._types.ResponseError: 404 page not found (status code: 404). What is the reason and how can I use it?

[dosubot[bot]]

To resolve the "ollama._types.ResponseError: 404 page not found" error when using the base_url with the ChatOllama model "gemma3:1b", ensure that the model is correctly pulled and available on your server. You can pull the model using the command ollama pull gemma3:1b. Additionally, make sure your Ollama Docker image is updated to the latest version to avoid any version incompatibility issues between the Ollama server and client libraries ^[1].

Also, ensure that the base_url does not end with a slash, as this can sometimes cause issues with URL resolution ^[2].

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Bug Report | Other}

[lc345]

@dosu the model is correctly, i can use it in another way

[dosubot[bot]]

Ensure that the base_url is correctly formatted as a URL string, such as "http://*.*.*.*:8001", without any trailing slashes or additional path segments. The expected format for the base_url parameter in the ChatOllama class is a URL string like "https://ollama-hostname:8000" ^[1]. Double-check that the server is configured to handle requests at the specified endpoint and that there are no network issues preventing access.

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Bug Report | Other}