The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →gpustat is a Python command-line utility for checking and repeatedly monitoring NVIDIA GPU status. Run gpustat for a one-time summary, or use gpustat --watch (also -i) to refresh it. Its compact output can include each GPU’s identity, temperature, utilization, memory use, and running processes.
What gpustat shows
The project describes gpustat as “A simple command-line utility for querying and monitoring GPU status.” A typical output row can show a GPU index and name, temperature, utilization percentage, used and total memory, and a process using GPU memory. The README’s example illustrates available fields; it does not guarantee that every field will appear for every device, driver, or environment. See the gpustat README for the project’s documentation.
gpustat is for NVIDIA graphics devices only; the upstream README says it does not support AMD GPUs. It is a status viewer, not a universal GPU monitoring tool.
Install gpustat
The upstream README documents installation with Python’s package installer:
#1 Best Overall
- AI Performance: 767 AI TOPS
- OC mode: 2632 MHz (OC mode)/ 2602 MHz (Default mode)
- Powered by the NVIDIA Blackwell architecture and DLSS 4
- Axial-tech fan design features a smaller fan hub that facilitates longer blades and a barrier ring that increases downward air pressure
- A 2.5-slot design maximizes compatibility and cooling efficiency for superior performance in small chassis
pip install gpustatCompatibility depends on the gpustat release and the installed Python, NVIDIA driver, and NVML Python binding. The README states that gpustat 1.2 and later requires nvidia-ml-py >= 12.535.108. It recommends nvidia-ml-py as the official NVML binding and warns against installing the separate pynvml package. Older gpustat releases have their own documented requirements, so check the README and PyPI package listing for the version you intend to install.
Run a one-time query or keep monitoring
Print the current status
After installation, run this in a terminal:
gpustatThis prints a snapshot of the GPUs visible to gpustat and the status information available from the system.
Rank #2
- Powered by the NVIDIA Blackwell architecture and DLSS 4
- Powered by GeForce RTX 5070 Ti
- Integrated with 16GB GDDR7 256bit memory interface
- PCIe 5.0
- WINDFORCE cooling system
Refresh the status repeatedly
Use --watch or its short form -i to refresh the display. The interval option controls how often it updates; consult the README’s command help for the exact syntax supported by your installed version.
gpustat --watchFor example, to set an interval explicitly, use the documented interval option:
Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Repair Windows errors before they cause bigger problems3Fix the driver behind crashes, sound loss and screen glitchesRank #3
- Powered by the NVIDIA Blackwell architecture and DLSS 4. System Requirements: Minimum 850W PSU with 16-pin 12V-2x6 (12VHPWR) connector required. Verify before purchasing.
- Military-grade components deliver rock-solid power and longer lifespan for ultimate durability. Compatibility: 348mm (13.7") length, 3.6 slots, 4.3 lbs. Confirm case clearance and slot spacing. GPU bracket included.
- Protective PCB coating helps protect against short circuits caused by moisture, dust, or debris
- 3.6-slot design with massive fin array optimized for airflow from three Axial-tech fans
- Phase-change GPU thermal pad helps ensure optimal thermal performance and longevity, outlasting traditional thermal paste for graphics cards under heavy loads
gpustat --watch --interval 2Repeated output is useful when observing changing utilization or memory use during a workload; it remains a live status view, not a performance benchmark.
Choose output and device details
gpustat includes switches for tailoring what it displays. The README documents JSON output, GPU selection, suppressing process details, and additional fields such as user, command, PID, fan, codec, and power information.
Rank #4
- Powered by the NVIDIA Blackwell architecture and DLSS 4
- Powered by GeForce RTX 5060
- Integrated with 8GB GDDR7 128bit memory interface
- PCIe 5.0
- WINDFORCE cooling system
| Need | Option | Use |
|---|---|---|
| Structured output for a script | --json |
Emit JSON rather than the standard formatted display. |
| Select GPU indices | --id |
Limit the query to specified GPU indices; use the syntax in your installed version’s help. |
| Hide process details | --no-processes |
Suppress process information in the display. |
| Show more context | Additional display switches | The README lists options for user, command, PID, fan, codec, and power details. |
Which fields are populated depends on the device and environment. For a Python program, the project also exposes gpustat.new_query(); results can be printed in formatted form or converted to a dictionary with jsonify().
Understand GPU index differences with CUDA
gpustat and nvidia-smi number GPUs by PCI bus ID, but CUDA may order devices differently. A row labeled GPU 0 in gpustat therefore does not necessarily correspond to CUDA device 0 in an application.
Best Value
- Powered by the NVIDIA Blackwell architecture and DLSS 4 OC mode: 2640MHz/Default mode: 2610MHz (Boost Clock)
- Military-grade components deliver rock-solid power and longer lifespan for ultimate durability
- Protective PCB coating helps protect against short circuits caused by moisture, dust, or debris
- 3.125-slot design with massive fin array optimized for airflow from three Axial-tech fans
- Phase-change GPU thermal pad helps ensure optimal thermal performance and longevity, outlasting traditional thermal paste for graphics cards under heavy loads
When you need CUDA and the status tools to use PCI bus order, the project recommends setting CUDA_DEVICE_ORDER=PCI_BUS_ID before setting CUDA_VISIBLE_DEVICES. Set these environment variables in the process environment before starting the CUDA application; CUDA_VISIBLE_DEVICES controls which devices that application sees.
When to use gpustat instead of nvidia-smi
Both tools can help inspect NVIDIA GPUs, but their interfaces are not identical. gpustat is positioned by its project as a compact status display. NVIDIA’s nvidia-smi documentation covers selective property queries, compute-application queries, monitoring commands, and broader management capabilities.
- Choose gpustat when its compact summary or its JSON output fits your terminal or script workflow.
- Choose
nvidia-smiwhen you need NVIDIA’s documented property or compute-process queries, monitoring options, or management commands. - Check the exact fields and output format your workflow needs; do not assume a command or field from one tool maps directly to the other.
The gpustat README asks “Just less than nvidia-smi?” as a project comparison prompt. That is not an independent performance comparison or evidence that one tool is universally better.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →
