Internet Speed Test FAQ

SpeedOf.Me is a free, browser-based internet speed test that measures your download speed, upload speed, latency, and jitter. Launched in 2011 as the first HTML5 speed test, it runs in any modern browser on any device, with no app or download required.

Visit SpeedOf.Me and click the START TEST button. The test measures your download speed, upload speed, latency, and jitter in under a minute. No app is needed, and it runs directly in your browser on any device.

Yes, SpeedOf.Me is completely free for personal use. Just visit the website and click START TEST. No account, registration, or app download is required. For developers and businesses, SpeedOf.Me also offers an API with free and paid plans to embed speed testing into applications.

SpeedOf.Me uses HTML5-based single-stream testing that reflects real-world browsing performance. Unlike speed tests that require an app, SpeedOf.Me runs in any browser on any device. It also uses globally distributed test servers and measures download, upload, latency, and jitter in a single test.

A good internet speed depends on your usage. For basic browsing and email, 10 Mbps is sufficient. For HD video streaming, 25 Mbps is recommended. For 4K streaming or online gaming, 50 to 100 Mbps provides a smooth experience. Households with multiple users or devices should aim for higher speeds.

For one person browsing and emailing, 10 to 25 Mbps is plenty. HD streaming needs about 5 to 10 Mbps per stream and 4K about 25 Mbps. Online gaming depends less on raw speed and more on low latency and jitter. For working from home with video calls, aim for at least 25 Mbps download and 5 to 10 Mbps upload. Add headroom for each additional person or device on the network.

Mbps means megabits per second; MBps means megabytes per second. There are 8 bits in a byte, so divide by 8 to convert: a 100 Mbps connection downloads at about 12.5 MBps. Internet plans and speed tests (including SpeedOf.Me) are measured in Mbps, while file sizes are usually shown in megabytes (MB).

Common causes of slow internet include Wi-Fi interference, network congestion during peak hours, too many devices on your network, outdated router firmware, distance from your router, or your ISP throttling speeds. Run a SpeedOf.Me test to check your actual speeds, then compare them with what your ISP plan promises.

Wi-Fi speed is reduced by distance from the router, walls, interference from nearby networks, and older Wi-Fi hardware, so it is often lower than your plan's rated speed. For the most accurate picture of what your ISP delivers, test on a device connected by Ethernet. Then test over Wi-Fi to see what your wireless connection actually provides where you use it.

If the test does not start or finish, an ad blocker, browser extension, VPN, or firewall is the most common cause, because they can block the test's requests. Try disabling your ad blocker for SpeedOf.Me, pausing extensions, or testing in a private or incognito window. Clearing your browser cache can also help.

Upload failures are usually caused by firewall or security software, a VPN, or a proxy that blocks outbound test data. Try temporarily disabling that software, switching networks, or using a different browser to confirm.

SpeedOf.Me uses single-stream testing, which measures your actual browsing and streaming speed rather than the theoretical maximum of your connection. This gives you a realistic picture of your everyday internet performance. Multi-stream tests can inflate results by saturating your connection in ways that don't reflect real usage.

Download speed measures how fast data travels from the internet to your device, which affects loading websites, streaming video, and downloading files. Upload speed measures how fast data travels from your device to the internet, which affects video calls, uploading files, and live streaming. Most internet plans offer faster download speeds than upload speeds.

Latency (ping) is the round-trip time for data between your device and a server, in milliseconds, where lower is more responsive. Jitter is the variation in latency over time; high jitter causes buffering on video calls and lag in games. SpeedOf.Me measures latency 10 times and reports the lowest, then reports jitter as the variation across those measurements. As a rough guide, jitter under 5 ms is excellent, under 20 ms is very good, and under 50 ms is good.

SpeedOf.Me automatically selects from a global network of over 130 points of presence (PoPs) at major internet exchanges. It picks the fastest, most reliable server for your connection, which is not always the physically closest one, so results reflect real-world performance rather than an idealized lab number. See how it works for more.

During the test, the real-time graph updates roughly every half second, showing loaded data divided by elapsed time. SpeedOf.Me downloads progressively larger sample files (up to 128 MB) until one takes long enough to measure accurately, and your final result is based on that largest completed sample rather than a simple average of the live readings.

Yes. SpeedOf.Me's test servers are IPv6-enabled, so if your connection supports IPv6 the test will use it automatically.

SpeedOf.Me works on virtually any device with a modern web browser: desktop computers, laptops, smartphones, tablets, smart TVs, game consoles (PlayStation, Xbox, Nintendo Switch), and even in-car infotainment systems. No app installation is required.

SpeedOf.Me works in any modern browser, including current versions of Chrome, Safari, Firefox, and Edge, on desktop and mobile. It just needs JavaScript enabled.

Your test history is stored locally in your own browser (localStorage), not on SpeedOf.Me's servers. You can export it to CSV from the History view, and clear it any time through your browser's site-data settings.

Yes. SpeedOf.Me publishes a free MCP (Model Context Protocol) server, @speedofme/mcp, that lets AI assistants run real speed tests for you. It works with Claude Desktop, Claude Code, Cursor, Windsurf, VS Code, Zed, and other MCP-compatible clients.

The MCP server connects AI agents to SpeedOf.Me. Add it to your AI client with a one-line command, npx -y @speedofme/mcp, and the assistant gains two tools: run_speed_test (returns download, upload, latency, jitter, the test server, and a timestamp) and get_test_history (returns your recent results). See the package on npm or the MCP server docs to get started. Not using an AI agent? The same speed tests are available via the CLI for scripts and automation.

Yes. With the MCP server connected, you can ask an AI assistant to run a speed test and interpret the results. For example, it can explain whether your latency and jitter are good enough for video calls or gaming, compare your measured speed against your ISP plan, or run repeated tests to spot patterns. The assistant uses the live run_speed_test and get_test_history data to reason about your connection.

Yes. The SpeedOf.Me API lets you embed accurate download, upload, latency, and jitter testing into websites, apps, and devices without hosting any test infrastructure. You can integrate with the JavaScript SDK, the MCP server for AI agents, or the CLI for scripts and automation.

The full API reference is at speedof.me/api/docs/.

Yes. Ready-to-use examples are at speedof.me/api/examples/, covering the web (vanilla JavaScript, React, Vue, Angular), mobile (React Native, Flutter, Android, iOS), and desktop (Electron, macOS, Windows).

Yes, you can start with a free trial. Paid plans scale from small projects to high-volume and enterprise use. See our plans and pricing page for current tiers and rates.

Yes. The API supports white-label and customizable speed-test widgets, real-time progress callbacks, and historical result storage, so the test can match your product's look and workflow.

Anything that runs JavaScript: websites, web apps, mobile apps (iOS, Android, React Native, Flutter), desktop apps (Electron, macOS, Windows), IoT devices, and AI agents via MCP (Model Context Protocol).

SpeedOf.Me offers a free sample bandwidth dataset on the AWS Marketplace, and API customers get an analytics dashboard that breaks their own test traffic down by country, city, ISP, and device. For custom aggregate datasets, such as bandwidth trends for a specific ISP, region, or mobile carrier, ask about an enterprise data plan on our API page.

Yes. API customers get a real-time analytics dashboard showing geographic distribution (country and city), device and speed insights, and usage statistics over 7, 30, or 90 days, with CSV export of the underlying rows.

Yes. On the website you can export your personal test history to CSV from the History view. API customers can additionally export their analytics data (per date, city, and test server) to CSV from the dashboard.