`/model` support for ollama/lmstudio
Resolved 💬 2 comments Opened Dec 19, 2025 by scaryrawr Closed May 10, 2026
💡 Likely answer: A maintainer (etraut-openai, contributor)
responded on this thread — see the highlighted reply below.
What feature would you like to see?
The /model slash command always recommends cloud based models even when the model provider is configured to ollama or lmstudio.
It's very useful to be able to switch to different local models depending on the task at hand.
The lmstudio and ollama clients have a fetch_models that lists the available models that could possibly be integrated?
Additional information
_No response_
2 Comments
Would need that for llama.cpp (
llama-server) too.According to endpoints in the sources https://github.com/ggml-org/llama.cpp/blob/a89002f07b55dace8671fc07b2e2418700716992/tools/server/README.md?plain=1#L1107 there is not a
fetch_modelsendpoint but a/v1/models. Stating that isOpenAI-compatible Model Info API. I am not sure.This feature request hasn't received any upvotes in several months, so I'm going to close it.