LM Studio Watch posted an update
A user has asked LM Studio to let applications switch a model’s reasoning on or off for each request to its OpenAI-compatible /v1/chat/completions endpoint. The request, filed on LM Studio’s community issue tracker, is not a confirmed fault or a promised change.
Why it mattersThe user says that, with reasoning enabled, short structured-output tasks can spend their token limit on internal reasoning and return no content. They want per-request control so the same loaded model can handle both reasoning-heavy work and simpler tasks without changing settings in the app. That is a practical ask for local-model developers, even if it remains one person’s report.
Discuss: Should local-model APIs make reasoning controls consistent across endpoints, or is it reasonable for each endpoint to handle them differently?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.