LM Studio 0.3.14: Multi-GPU Controls Released
Action Required
Users with multi-GPU systems can now optimize model performance and memory usage, leading to faster inference speeds and reduced memory pressure.
AI Impact Summary
LM Studio 0.3.14 introduces significant improvements for users with multi-GPU systems, offering granular control over GPU allocation and memory management. This includes the ability to enable/disable individual GPUs, choose allocation strategies, and limit model weights to dedicated GPU memory, addressing potential performance bottlenecks and memory constraints. These features cater to advanced users and developers seeking to optimize model performance on multi-GPU setups, particularly those with NVIDIA GPUs.
Affected Systems
- Date
- Date not specified
- Change type
- capability
- Severity
- high