intel-pti

What it is and what it does

Intel PTI is a framework for developing custom profiling and analysis tools for applications built on Intel's oneAPI programming model and running on Intel accelerators (XPUs). It sits between your application and the Intel hardware, exposing instrumentation hooks that let you capture performance data, trace execution, and analyze workload behavior.

The package is positioned for developers, researchers, and system administrators who need to understand or optimize oneAPI workloads on Intel accelerators. It has no runtime dependencies, but it is tightly coupled to Intel's hardware ecosystem and oneAPI stack—you'll need compatible hardware or a oneAPI development environment to use it effectively. The framework is actively maintained and carries a permissive MIT license.

Use it for:

• Build custom profiling tools to measure performance of oneAPI applications on Intel GPUs or other XPUs.
• Trace and analyze execution patterns of accelerated workloads to identify bottlenecks.
• Instrument scientific or AI applications running on Intel accelerators for detailed performance telemetry.
• Develop monitoring solutions for oneAPI-based HPC or data-center workloads.
• Analyze memory access patterns and kernel execution on Intel XPU hardware.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Provides a framework for building profiling tools that instrument and analyze applications built on oneAPI and running on Intel accelerators (XPUs).

Yes, if you develop or optimize oneAPI applications on Intel accelerators and need a profiling framework. The permissive license, active maintenance, and zero runtime dependencies make it low-friction to add. No, if you lack Intel XPU hardware or work outside the oneAPI ecosystem—the package is tightly specialized and will not be useful otherwise.

Install

intel-pti on PyPI

pip install intel-pti

uv add intel-pti

poetry add intel-pti

Installing intel-pti

Before you install

Medium install friction due to platform-specific wheels (Linux x86_64 and Windows x64 only); actively maintained with a recent release.

License in practice

Licensed under MIT, a permissive open-source license that allows use, modification, and distribution with minimal restrictions.

Quickstart

pip install intel-pti
import intel_pti
# Use the framework to build profiling tools for oneAPI applications

Requires Intel XPU hardware or compatible oneAPI environment; wheels available only for Linux x86_64 and Windows x64.

Verify before relying

• Specific Python version requirements (requires_python is unspecified in metadata)
• Whether this package provides ready-to-use profiling tools or requires substantial custom development
• Compatibility with specific oneAPI versions or Intel XPU models

Package facts

Evidence: intel_pti-1.0.1-py2.py3-none-manylinux_2_28_x86_64.whl; intel_pti-1.0.1-py2.py3-none-win_amd64.whl

Tags

More Libraries packages