Rate this Page

MPS Environment Variables#

Created On: Jun 11, 2025 | Last Updated On: Jun 11, 2025

PyTorch Environment Variables

Variable

Description

PYTORCH_DEBUG_MPS_ALLOCATOR

If set to1, set allocator logging level to verbose.

PYTORCH_MPS_LOG_PROFILE_INFO

Set log options bitmask toMPSProfiler. SeeLogOptions enum inaten/src/ATen/mps/MPSProfiler.h.

PYTORCH_MPS_TRACE_SIGNPOSTS

Set profile and signpost bitmasks toMPSProfiler. SeeProfileOptions andSignpostTypes.

PYTORCH_MPS_HIGH_WATERMARK_RATIO

High watermark ratio for MPS allocator. Default is 1.7.

PYTORCH_MPS_LOW_WATERMARK_RATIO

Low watermark ratio for MPS allocator. Default is 1.4 (unified) or 1.0 (discrete).

PYTORCH_MPS_FAST_MATH

If1, enables fast math for MPS kernels. See section 1.6.3 in theMetal Shading Language Spec.

PYTORCH_MPS_PREFER_METAL

If1, uses metal kernels instead of MPS Graph APIs. Used for matmul.

PYTORCH_ENABLE_MPS_FALLBACK

If1, falls back to CPU when MPS ops aren’t supported.

Note

high watermark ratio is a hard limit for the total allowed allocations

  • 0.0 : disables high watermark limit (may cause system failure if system-wide OOM occurs)

  • 1.0 : recommended maximum allocation size (i.e., device.recommendedMaxWorkingSetSize)

  • >1.0: allows limits beyond the device.recommendedMaxWorkingSetSize

e.g., value 0.95 means we allocate up to 95% of recommended maximumallocation size; beyond that, the allocations would fail with OOM error.

low watermark ratio is a soft limit to attempt limiting memory allocations up to the lower watermarklevel by garbage collection or committing command buffers more frequently (a.k.a, adaptive commit).Value between 0 to m_high_watermark_ratio (setting 0.0 disables adaptive commit and garbage collection)e.g., value 0.9 means we ‘attempt’ to limit allocations up to 90% of recommended maximumallocation size.