MANDUMON
Web Bench – Online Performance Test — CPU, GPU & Memory Benchmark, No Install
Web Bench uses 8 categories and dozens of detailed tests to see how your device performs inside the browser. Tests your device or browser does not support are skipped automatically, and every value is a reference that varies by environment. Choose a quick test (about 30 seconds), standard test (2–3 minutes) or 5/10 minute stress test to check your CPU, GPU and memory speed right in the browser.
Use it nowFeatures
CPU single-core · CPU multicore
CPU Single measures how fast one core finishes real-world work such as text processing, JSON parsing, compression, hashing, image processing, sorting, pathfinding, physics simulation, ray tracing and linear algebra. CPU Multi runs the same kind of work split across several Workers to measure throughput when cores work together.
Memory · Browser · Storage
Memory looks at the throughput of reading, writing and copying a contiguous 16 MB block, along with the latency of chasing data through random access. The Browser category measures DOM manipulation and layout, Canvas 2D drawing, screen refresh rate and frame consistency. Storage creates a temporary file in the browser's OPFS to measure sequential write, sequential read, random write and random read speed, then deletes the file when finished.
GPU graphics · GPU compute
GPU Graphics renders large numbers of triangles, 131,072 particles and bloom and blur post-processing off-screen and measures the average FPS. If WebGPU is unavailable, the test may run on the WebGL2 path instead, and if neither is available this category is skipped. GPU Compute measures speed when you hand calculations, not graphics, to the GPU. Without WebGPU or the shader-f16 feature, the affected tests are skipped.
AI inference (optional)
AI inference is optional and runs only if you agree to download the models. It feeds fixed test sentences and images to measure how fast text generation and image models run inside the browser, and it never uses your personal files. It is a reference for how smooth on-device AI features might feel, and a higher processing speed is better. If you do not agree, it is not run and is left out of the total score. If WebGPU is unavailable it runs on the WASM path instead, and that is shown separately.
Browser battle
Share a result and open its link in another browser on the same device. Select Benchmark this browser and compare to use the same benchmark mode. Compare categories, detailed values and environments in the battle card, keeping power and thermal conditions consistent.
Rankings
Only full test results you agree to publish enter the rankings. Devices are grouped by detected GPU name and core count; typed model names are never sent. Compare device medians from full tests using the same score formula version. Devices with at least 3 samples are ranked; those with 1–2 appear separately. Updates may take 60 seconds.
How to use
- On the test screen, choose a quick test, a standard test or a stress test.
- Press Start and keep the tab in front until the test finishes.
- Check your score and per-category results, then save a PNG card or compare with your last 20 results.
Use cases
- Check performance in your current browser on a PC or phone without installing software.
- Compare power modes on the same device, or use a stress test to observe performance under sustained load.
- Switch browsers on the same device to compare scores and execution environments side by side.
Frequently asked questions
Do I need an installation or login?
No. Run the tests in a PC or phone browser without installing software or logging in.
Is AI testing required?
No. It runs only with consent to download the models. If you skip it, it is excluded from testing and the overall score.
Where are results stored?
Measurements run on your device. The latest 20 results stay in this browser and can be removed by clearing site data. Share results as PNG cards or links. Results are sent to the server only when you agree to ranking registration.
Why do scores differ between browsers?
Browsers use different JavaScript engines, graphics APIs and security policies, so scores vary even on the same device. Start a browser battle from a shared link to compare detailed measurements and execution environments.
Related guides
How to Measure Laptop and Phone Performance Accurately: A Power, Heat and Background App Checklist
Why does the same device score differently every time? Power mode, thermal throttling and background apps are the usual causes. Get a pre-test checklist, learn when to use the quick, standard or stress test, and how to compare runs.
Can Your Browser Run AI? Checking WebGPU and On-Device AI Performance
What does it take to run AI inside a browser? Learn how WebGPU differs from WebGL, the benefits of on-device AI, shader-f16 and the WASM fallback, and how to read Web Bench's optional AI inference and GPU compute results.
How to Read Your Web Bench Score: 8 Categories and Single vs Multi-Core
What the Web Bench total and its eight categories measure, why CPU multi-core scores run so much higher than single-core, how to read strength and weakness labels and the bar chart, and what to watch when comparing devices.
Why the Same Device Scores Differently in Each Browser: How to Compare Chrome, Safari, Edge and Firefox
Learn why benchmark scores differ between browsers on the same device, from JavaScript engines and WebGPU to security policies and extensions, and how to compare them fairly with Web Bench's Browser Battle.