Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Nvidia Optimus Laptops. LM Studio at idle(with unloaded models) prevents the discrete GPU from entering power saving mode #450

Open
ThatUsernameAlreadyExist opened this issue Feb 20, 2025 · 0 comments

Comments

@ThatUsernameAlreadyExist

Which version of LM Studio?
LM Studio 0.3.10 (build 6) with CUDA llama.cpp runtime

Which operating system?
Windows 11 24H2

What is the bug?
LM Studio is running in the background to work on demand(Local LLM Service (headless) with JIT mode), and all LLM models have been unloaded from memory (meaning the GPU is not being used). However, on a laptop with Nvidia Optimus, LM Studio does not allow the discrete Nvidia GPU to enter power-saving mode (it continues to use it). This leads to increased heating and faster battery drain. On the previous version, 0.3.9, this issue did not exist; the discrete GPU went into sleep immediately after the models were unloaded from memory.
The problem in the new version may be due to a constantly running additional process "LM Studio.exe", which keeps CUDA libraries loaded.

Screenshots
Not sleeping Nvidia GPU:
Image

LM Studio process that used CUDA libs at idle:
Image

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants