---
id: intel-sycl-rt
version: "2026.1.1"
license: Intel End User License Agreement for Developer Tools
license_treatment: unclear
maintenance: active
---

# intel-sycl-rt — Intel® oneAPI DPC++/C++ SYCL Compiler Runtime package

License: unclear · Maintenance: active · Downloads: 232.7K/mo

## What it is and what it does

intel-sycl-rt is a compiler-specific runtime package for the Intel oneAPI DPC++/C++ SYCL compiler. It contains the shared libraries needed to execute binaries that were compiled with the Intel SYCL compiler, allowing deployment to machines that do not have the full oneAPI development toolkit installed. The package is part of Intel's oneAPI ecosystem, which aims to provide a unified programming model for heterogeneous computing across multiple architectures and vendors.

This is a deployment-focused package: if you have already compiled code with the Intel DPC++ compiler, you use intel-sycl-rt to run that code on target systems. It is not for writing or compiling SYCL code—that requires the full development toolkit. The package depends on intel-cmplr-lic-rt, intel-cmplr-lib-rt, and intel-cmplr-lib-ur, and according to its description also requires separate common libraries and licensing packages to function completely.

Use it for:

- Deploy SYCL-compiled executables to production servers without installing the full Intel oneAPI development toolkit
- Distribute compiled binaries that use Intel DPC++ to end users or other teams
- Set up minimal runtime environments for running heterogeneous compute workloads on Intel GPUs or CPUs
- Create Docker or container images with only the runtime dependencies needed for SYCL executables

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Provides the Intel oneAPI DPC++/C++ SYCL compiler runtime libraries needed to run executables compiled with the Intel SYCL compiler on systems without the full development toolkit installed.

Yes, if you have executables compiled with the Intel DPC++ SYCL compiler and need to run them on systems without the full development toolkit. The package is actively maintained and has no known vulnerabilities. However, verify that all three required Intel compiler runtime dependencies are available in your environment, and review the Intel End User License Agreement terms before deploying in production.

## Install

pip install intel-sycl-rt
uv add intel-sycl-rt
poetry add intel-sycl-rt

## Installing intel-sycl-rt

Before you install: Medium install friction due to three Intel compiler runtime dependencies (intel-cmplr-lic-rt, intel-cmplr-lib-rt, intel-cmplr-lib-ur) that must also be present. Package is actively maintained with a recent release.

License in practice: Licensed under Intel End User License Agreement for Developer Tools with unclear treatment—not an open-source license. Review the terms before deploying in production or redistributing compiled binaries.

Quickstart:

pip install intel-sycl-rt
# Requires intel-cmplr-lic-rt, intel-cmplr-lib-rt, intel-cmplr-lib-ur
# to be installed for runtime execution of compiled SYCL code

Requires the three Intel compiler runtime packages (intel-cmplr-lic-rt, intel-cmplr-lib-rt, intel-cmplr-lib-ur) to be installed. Only available for Linux (manylinux_2_28_x86_64) and Windows (win_amd64).

Verify before relying:

- Whether the three runtime dependencies are automatically pulled in or must be manually installed
- Exact Python version compatibility (requires_python is unspecified in metadata)
- Whether this package alone is sufficient or if additional oneAPI common libraries are always required

## Package facts

- License: Intel End User License Agreement for Developer Tools (unclear)
- Python support: unspecified
- Install friction: medium
- Maintenance: active
- Downloads: 232.7K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

sycl runtime deployment, intel dpc++ runtime libraries, oneapi compiler runtime, sycl executable deployment, intel compiler shared libraries, compiler-runtime, heterogeneous-compute, intel-oneapi

[View on SkillFed](https://skillfed.io/packages/intel-sycl-rt) · [View on PyPI](https://pypi.org/project/intel-sycl-rt/)