cuda.core.system.Temperature#

class cuda.core.system.Temperature(int handle: int)#

Methods

__init__(*args, **kwargs)#
sensor(
self,
sensor: TemperatureSensors = TemperatureSensors.TEMPERATURE_GPU,
) int#

Get the temperature reading from a specific sensor on the device, in degrees Celsius.

Parameters:

sensor (TemperatureSensors, optional) – The temperature sensor to query.

Returns:

The temperature in degrees Celsius.

Return type:

int

thermal_settings(
self,
sensor_index: ThermalTarget,
) ThermalSettings#

Used to execute a list of thermal system instructions.

TODO: The above docstring is from the NVML header, but it doesn’t seem to make sense.

Parameters:

sensor_index (ThermalTarget) – The index of the thermal sensor.

Returns:

The thermal settings for the specified sensor.

Return type:

ThermalSettings

threshold(
self,
threshold_type: TemperatureThresholds,
) int#

Retrieves the temperature threshold for this GPU with the specified threshold type, in degrees Celsius.

For Kepler™ or newer fully supported devices.

See TemperatureThresholds for possible threshold types.

Note: This API is no longer the preferred interface for retrieving the following temperature thresholds on Ada and later architectures: NVML_TEMPERATURE_THRESHOLD_SHUTDOWN, NVML_TEMPERATURE_THRESHOLD_SLOWDOWN, NVML_TEMPERATURE_THRESHOLD_MEM_MAX and NVML_TEMPERATURE_THRESHOLD_GPU_MAX.

Support for reading these temperature thresholds for Ada and later architectures would be removed from this API in future releases. Please use get_field_values() with NVML_FI_DEV_TEMPERATURE_* fields to retrieve temperature thresholds on these architectures.

Attributes

margin#

int

The thermal margin temperature (distance to nearest slowdown threshold) for the device.

Type:

Temperature.margin