Skip to content
SPECIFICTION
Get listed

StreamBench

github.com

WindowsmacOSLinux
Visit github.com
Spec score 7.2 #58 of 73 in PC Benchmark Software

Spec sheet

Purpose
StreamBench benchmarks memory bandwidth and AI inference across CPU, GPU, and NPU hardware.Read 11 Oct 2026 · github.com
Intended users
The README describes it as aimed at users seeking a quick visual benchmark with prebuilt binaries, GPU coverage, and AI inference comparisons on modern hardware.Read 11 Oct 2026 · github.com
Platforms
Prebuilt downloads are offered for Windows and macOS, while the project also documents Linux use and source builds.Read 11 Oct 2026 · github.com
Privacy
Saved results and CLI summaries may include hardware details and the hostname, and the project advises reviewing files before sharing them publicly.Read 11 Oct 2026 · github.com
AI backends
The AI inference benchmark supports Microsoft Foundry Local, LM Studio, and Ollama.Read 11 Oct 2026 · github.com
AI integrations
It supports Microsoft Foundry Local, LM Studio, and Ollama as AI backends.Read 11 Oct 2026 · github.com
AI measurements
The AI benchmark reports model load time, first-run response time, and warm-run tokens per second.Read 11 Oct 2026 · github.com
Array size
The README recommends using arrays at least four times larger than the largest cache and notes that small sizes can be skewed by cache effects.Read 11 Oct 2026 · github.com
Automation
Range testing can sweep array sizes and save consolidated CSV output.Read 11 Oct 2026 · github.com
CLI or automation
YesRead 22 Sep 2026 · github.com
Custom test profiles
YesRead 22 Sep 2026 · github.com
Device coverage
Foundry Local supports CPU, GPU, and NPU benchmarking; LM Studio and Ollama do not support NPU benchmarking.Read 11 Oct 2026 · github.com
Downloads
The project offers prebuilt binaries for Windows and macOS, with x64 and ARM64 variants.Read 11 Oct 2026 · github.com
GPU testing
The GPU memory bandwidth backend uses OpenCL, which the project says is loaded dynamically without a GPU SDK dependency.Read 11 Oct 2026 · github.com
Limits
The project is focused on single-node benchmarking and does not include the MPI support listed for upstream STREAM.Read 11 Oct 2026 · github.com
Linux support
The README documents macOS and Linux launch scripts and says the project can be built from source on Linux.Read 11 Oct 2026 · github.com
macOS limitation
Foundry Local on macOS currently supports GPU mode only, without CPU or NPU device targeting.Read 11 Oct 2026 · github.com
Measured AI metrics
It reports model load time, first-response time, and warm-run tokens per second.Read 11 Oct 2026 · github.com
Memory tests
Its CPU and GPU benchmarks run the Copy, Scale, Add, and Triad STREAM kernels.Read 11 Oct 2026 · github.com
NPU limitation
NPU benchmarking requires Foundry Local; the README says LM Studio and Ollama do not support NPU benchmarking.Read 11 Oct 2026 · github.com
Operating systems
Windows; macOS; LinuxRead 22 Sep 2026 · github.com
Requirements
AI benchmarking requires at least one supported AI backend, which the setup script can install.Read 11 Oct 2026 · github.com
Result export formats
JSON; CSVRead 22 Sep 2026 · github.com
Results
The CLI displays benchmark results and can save them in JSON and CSV formats.Read 11 Oct 2026 · github.com
Saved results
The CLI can save benchmark results in JSON and CSV formats.Read 11 Oct 2026 · github.com
Workload areas
CPU memory bandwidth; GPU memory bandwidth; AI inferenceRead 22 Sep 2026 · github.com

Where Specifiction places it

Alternatives to StreamBenchAll alternatives →