clpeak

At a glance

• Editor score
  3.8 / 10
• Pricing
  Open source
• Best for
  Cross-platform compute and memory-throughput testing
• Paid from
  Free
• Workload areas
  compute throughput; memory bandwidth; matrix throughput; kernel launch latency; AI inference
• Facts checked
  23 Sep 2026

• Where it wins

  • Measures compute, memory bandwidth, matrix throughput and kernel launch latency.
  • Supports Windows, macOS, Linux, iOS and Android.
  • Exports results as JSON and compares runs with saved result files.

• Where it doesn't

  • Has no comparison database for benchmarking against a broader result pool.
  • Synthetic benchmarks do not provide a broad measure of real-world application performance.
  • Saved-result comparisons require users to supply their own baseline files.

Our verdict on clpeak

clpeak is an open-source synthetic benchmark for measuring compute throughput, memory bandwidth, matrix performance and kernel launch latency on CPUs, GPUs and NPUs. It is aimed at solo users and small teams who want to examine hardware across Windows, macOS, Linux, iOS and Android. Its workload areas also include AI inference, and its command-line interface and Flutter desktop and mobile application offer different ways to run tests and review results. The tool is a focused fit for people comparing compute and memory behavior across devices or execution backends, rather than buyers seeking a general PC performance score.

Backend support is central to clpeak’s scope: listed options include OpenCL, Vulkan, CUDA, ROCm/HIP, Metal, oneAPI/SYCL, ONNX, Core ML and LiteRT. Multiple execution backends enable cross-backend comparisons, while device enumeration and selection help direct a run to a particular device. The benchmark also includes vendor-optimized GEMM tests alongside matrix benchmarks, plus memory-transfer tests and kernel launch latency measurement. This combination makes it useful for examining distinct throughput and latency characteristics; it does not turn synthetic results into a measure of performance in every application.

Results can be exported as JSON, and clpeak can compare a run against saved result files. That gives users a way to retain and revisit their own baselines, but there is no comparison database for placing results in a wider shared pool. As an open-source product, it does not present a paid tier structure in the listed offering. Choose clpeak if you want cross-platform, backend-aware synthetic testing and can manage your own comparisons. Choose another option if your decision depends on a built-in database of other systems’ results or application-focused performance evidence.

Your product? Our editors can test clpeak in depth and publish a review, labelled as paid · from $149/mo →

clpeak pricing

clpeak fact sheet

Alternatives to clpeak

• GGeekbenchA broad benchmark suite for comparing CPU and GPU performance across five platforms.9.3
• PassMark PerformanceTestA broad benchmark suite with free standard tests and a $39 one-time advanced tier.9.1
• 3DMarkA broad benchmark suite for comparing GPU, CPU, storage, and gaming performance.9.1

See all clpeak alternatives →

Featured on iTechGuides

clpeak is listed in our PC Benchmark Software directory. Add the badge to your site — it links back to this page.

<a href="https://www.itechguides.com/products/clpeak/"><img src="https://www.itechguides.com/best/badge/clpeak.svg" alt="Featured on iTechGuides" width="230" height="46"></a>

Reviewed by iTechGuides Editors · Editorial team · Updated Oct 2026

Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. Bought places stand at the number bought; this page and our disclosure say so. Paying vendors may get a fuller, more detailed review from our editors, and that review can change their score. How we rank.

Last updated 7 Oct 2026 · How we research and update