lmstudio-ai / lmstudio-ai/lmstudio-python

Loading a model through the SDK does not use per model defaults

Open
#178 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
875
Forks
166
PR merge metrics
No merged PRs in 30d

Description

Let’s take Qwen3 1.7B. The installed model default temp is 0.8.
I change the default temp to 0.6.
When I load the model in the UI, the temp is 0.8.
If I use the cli with lms load, the temp is 0.8.
If I use the REST API v1, the temp is 0.8
If I use the python sdk lmstudio.llm(), the temp is not 0.8, but back to 0.6.

The other params follow the same pattern - use my custom defaults when loaded everywhere except the python sdk.

The SDK should load using the per model custom defaults I set.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start at the Python SDK entry point lmstudio.llm() and trace how model-loading parameters are selected. Compare its handling of per-model custom defaults with the UI, CLI lms load, and REST API v1 behavior described in the issue. Done means SDK-loaded models use the configured per-model defaults, including temperature and the other parameters.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
api
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
45/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.