Just less than nvidia-smi?
NOTE: This works with NVIDIA Graphics Devices only, no AMD support as of now. Contributions are welcome!
Self-Promotion: A web interface of gpustat
is available (in alpha)! Check out gpustat-web.
Install from PyPI:
pip install gpustat
If you don't have root (sudo) privilege, please try installing gpustat
on user namespace: pip install --user gpustat
.
To install the latest version (master branch) via pip:
pip install git+https://github.com/wookayin/gpustat.git@master
gpustat
uses NVIDIA's official python bindings for NVML library (pynvml). As of now gpustat
requires nvidia-ml-py >= 11.450.129
, which is compatible with NVIDIA driver versions R450.00 or higher. Please upgrade the NVIDIA driver if gpustat
fails to display process information. If your NVIDIA driver is too old, you can use older gpustat
versions (pip install gpustat<1.0
). See #107 for more details.
- gpustat<1.0: Compatible with python 2.7 and >=3.4
- gpustat 1.0: Python >= 3.4
- gpustat 1.1: Python >= 3.6
$ gpustat
Options (Please see gpustat --help
for more details):
--color
: Force colored output (even when stdout is not a tty)--no-color
: Suppress colored output-u
,--show-user
: Display username of the process owner-c
,--show-cmd
: Display the process name-f
,--show-full-cmd
: Display full command and cpu stats of running process-p
,--show-pid
: Display PID of the process-F
,--show-fan
: Display GPU fan speed-e
,--show-codec
: Display encoder and/or decoder utilization-P
,--show-power
: Display GPU power usage and/or limit (draw
ordraw,limit
)-a
,--show-all
: Display all gpu properties above--id
: Target and query specific GPUs only with the specified indices (e.g.--id 0,1,2
)--no-processes
: Do not display process information (user, memory) (#133)--watch
,-i
,--interval
: Run in watch mode (equivalent towatch gpustat
) if given. Denotes interval between updates.--json
: JSON Output (#10)--print-completion (bash|zsh|tcsh)
: Print a shell completion script. See #131 for usage.
- Try
gpustat --debug
if something goes wrong. - To periodically watch, try
gpustat --watch
orgpustat -i
(#41).- For older versions, one may use
watch --color -n1.0 gpustat --color
.
- For older versions, one may use
- Running
nvidia-smi daemon
(root privilege required) will make querying GPUs much faster and use less CPU (#54). - The GPU ID (index) shown by
gpustat
(andnvidia-smi
) is PCI BUS ID, while CUDA uses a different ordering (assigns the fastest GPU with the lowest ID) by default. Therefore, in order to ensure CUDA andgpustat
use same GPU index, configure theCUDA_DEVICE_ORDER
environment variable toPCI_BUS_ID
(before settingCUDA_VISIBLE_DEVICES
for your CUDA program):export CUDA_DEVICE_ORDER=PCI_BUS_ID
.
[0] GeForce GTX Titan X | 77°C, 96 % | 11848 / 12287 MB | python/52046(11821M)
[0]
: GPU index (starts from 0) as PCI_BUS_IDGeForce GTX Titan X
: GPU name77°C
: GPU Temperature (in Celsius)96 %
: GPU Utilization11848 / 12287 MB
: GPU Memory Usage (Used / Total)python/...
: Running processes on GPU, owner/cmdline/PID (and their GPU memory usage)
See CHANGELOG.md