pyrsmi
Python System Management Library for AMD GPUs
Decision gist · record as of 2026-08-14
Yes, if you have AMD Instinct MI-series GPUs on a ROCm 6.0+ Linux system and need Python-based GPU monitoring or management. The package has no runtime dependencies, permissive licensing, and active maintenance. The aging status reflects time since release rather than abandonment. Install friction is low. Verify that your ROCm installation includes libamd_smi.so and consider installing the optional amdsmi package for full device naming support.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires ROCm 6.0 or later with libamd_smi.so library installed; Linux OS only; AMD Instinct MI-series GPU required; Python 3.9 or later.
- Low install friction with no runtime dependencies.
- Maintenance status is aging—last commit 2025-12-15 with 254 days since release—but the package is actively maintained on GitHub.
License · maintenance · safety
MIT License (permissive) — MIT License (permissive) allows unrestricted use, modification, and distribution in both open and closed-source projects with minimal obligations.
last release 2025-12-03 (254 days) · last repo commit 2025-12-15 · 25 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 116,328 downloads/mo, #12,210 on PyPI
Alternatives
Verify before relying
pip install pyrsmi
from pyrsmi import rocml
rocml.smi_initialize()
num_gpus = rocml.smi_get_device_count()
for i in range(num_gpus):
name = rocml.smi_get_device_name(i)
util = rocml.smi_get_device_utilization(i)
print(f"GPU {i}: {name}, Utilization: {util}%")
rocml.smi_shutdown()- Whether the amdsmi package is truly optional or required for core functionality beyond device naming.
- Performance improvement claims ('10-50x faster UUID retrieval') and backward compatibility scope.
- Thread-safety guarantees and behavior under concurrent access patterns.
- Support matrix for specific MI-series models beyond the tested MI350X, MI300X, MI250X, MI210.
What it is and what it does
pyrsmi is a Python wrapper around AMD's System Management Interface (SMI) library for monitoring and controlling AMD Instinct MI-series GPUs. It exposes functions to enumerate devices, query memory (total, used, reserved pages), measure utilization and power consumption, and retrieve device identifiers and metadata. Version 1.0+ migrated from the deprecated rocm-smi-lib to the modern amdsmi backend, maintaining backward-compatible APIs while adding access to PCIe metrics and extended telemetry.
The package is designed for system administrators, monitoring tools, and GPU-aware applications running on Linux with ROCm 6.0+. It requires explicit initialization and shutdown calls and returns -1 or empty strings on error. The optional amdsmi Python package is recommended for correct device naming and improved compatibility in containerized environments, though the core library functions work without it.
Use it for
- Monitor GPU utilization, memory usage, and power consumption in real-time dashboards or logging systems.
- Enumerate and identify AMD Instinct GPUs in a system for resource allocation and scheduling.
- Track memory allocation patterns and reserved pages for debugging GPU memory issues.
- Build automated thermal or power management tools that query fan RPM and average power draw.
- Integrate GPU telemetry into cluster management or containerized workload orchestration systems.
- Retrieve device UUIDs and identifiers in formats compatible with ROCm, raw, or NVIDIA conventions.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you have AMD Instinct MI-series GPUs on a ROCm 6.0+ Linux system and need Python-based GPU monitoring or management.
The package has no runtime dependencies, permissive licensing, and active maintenance. The aging status reflects time since release rather than abandonment. Install friction is low. Verify that your ROCm installation includes libamd_smi.so and consider installing the optional amdsmi package for full device naming support.
Install
pyrsmi on PyPI
Before you install
Low install friction with no runtime dependencies. Maintenance status is aging—last commit 2025-12-15 with 254 days since release—but the package is actively maintained on GitHub. Requires ROCm 6.0+ and the libamd_smi.so library to be present on the system.
Requires ROCm 6.0 or later with libamd_smi.so library installed; Linux OS only; AMD Instinct MI-series GPU required; Python 3.9 or later.
License in practice
MIT License (permissive) allows unrestricted use, modification, and distribution in both open and closed-source projects with minimal obligations.
Quickstart
pip install pyrsmi
from pyrsmi import rocml
rocml.smi_initialize()
num_gpus = rocml.smi_get_device_count()
for i in range(num_gpus):
name = rocml.smi_get_device_name(i)
util = rocml.smi_get_device_utilization(i)
print(f"GPU {i}: {name}, Utilization: {util}%")
rocml.smi_shutdown()Verify before relying
- Whether the amdsmi package is truly optional or required for core functionality beyond device naming.
- Performance improvement claims ('10-50x faster UUID retrieval') and backward compatibility scope.
- Thread-safety guarantees and behavior under concurrent access patterns.
- Support matrix for specific MI-series models beyond the tested MI350X, MI300X, MI250X, MI210.
Package facts
| License | MIT License permissive |
| Python support | Supports the current Python release >=3.9 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | None |
| Maintenance | Aging 254 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 116,328 / month, #12,210 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Development Status :: 3 - AlphaIntended Audience :: DevelopersIntended Audience :: System AdministratorsLicense :: OSI Approved :: MIT LicenseOperating System :: POSIX :: LinuxProgramming Language :: Python :: 3Programming Language :: Python :: 3 :: OnlyProgramming Language :: Python :: 3.10Programming Language :: Python :: 3.11Programming Language :: Python :: 3.12Programming Language :: Python :: 3.9Topic :: Software Development :: Libraries :: Python ModulesTopic :: System :: HardwareTopic :: System :: Systems Administration |
Evidence: pyrsmi-1.1.0-py2.py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “AMD GPU monitoring python”
- pyrsmiProvides Python bindings to monitor and manage AMD Instinct MI-series…
- hip-pythonProvides low-level Python and Cython bindings for HIP, enabling…
- torch-directmlEnables PyTorch to use DirectML for GPU acceleration on DirectX…
Give your agent the search over MCP, or paste the wish link into any chat.
More Python Modules packages
Converts domain names between Unicode and ASCII-compatible encoding (Punycode) according to IDNA 2008 and Unicode Technical Standard 46, with security validation and broader script coverage than the standard library.
Install it if you work with internationalized domain names, need to validate domains, or use HTTP clients that depend on it transitively.
Setuptools is a Python build backend and package management tool that handles building, distributing, and installing Python packages, including support for C/C++ extension modules.
PyYAML parses and emits YAML 1.1 data format, enabling serialization and deserialization of configuration files and Python objects to and from human-readable YAML text.
Pydantic validates Python data structures against type hints, coercing and checking input at runtime to ensure it matches a declared schema.
Provides reusable metadata objects for use with PEP-593 `typing.Annotated` to express common constraints like bounds, collection sizes, and predicates on types.
Install it if you use or build libraries that need to express type constraints in a standardized, inspectable way—or if you want to annotate your own types with…
Provides runtime tools to inspect and introspect Python type annotations, enabling programmatic examination of type hints at execution time.
See also hip-python · gpustat · GPUtil · py3nvml · flydsl · quadrants · nixl-cu13 · torch-directml · nvgpu · vastai