Python module
max.profiler.gpu
GPU diagnostics API.
This module allows accessing information about GPU(s) on the system. Information can be accessed synchronously or collected in the background for later retrieval.
Monitoring
GPUDiagContext | Context manager providing unified access to GPU diagnostic information across NVIDIA and AMD hardware. |
|---|---|
BackgroundRecorder | Asynchronous GPU metrics collection and data export capabilities. |
Stats
GPUStats | GPU state snapshot: memory, utilization, and clock metrics. |
|---|---|
MemoryStats | Detailed GPU memory usage statistics including total, free, used, and reserved memory. |
UtilizationStats | GPU compute and memory activity utilization percentages. |
ClockStats | Current and maximum GPU clock rates plus throttle reasons. |
ThrottleReason | alias of Literal['gpu_idle', 'applications_clocks_setting', 'sw_power_cap', 'hw_slowdown', 'sync_boost', 'sw_thermal_slowdown', 'hw_thermal_slowdown', 'hw_power_brake_slowdown', 'display_clock_setting'] |
HARDWARE_THROTTLE_REASONS | Build an immutable unordered collection of unique elements. |
|---|
Was this page helpful?
Thank you! We'll create more content like this.
Thank you for helping us improve!