Measure download, upload, ping, jitter and how your connection behaves under load, straight from your browser to the nearest of 300+ test locations worldwide.
Something blocked the connection to the test server.
Runs directly between your browser and the nearest test server. Results are never storedUses about 25 to 800 MB, depending on your speed
Upload speed decides how well you can share a local web app, game server, NAS or Home Assistant with the internet. Localtonet publishes them through a secure tunnel, with no port forwarding and even behind CGNAT.
| When | Download | Upload | Ping | Jitter | Server |
|---|
Stored only in this browser.
The test runs entirely in your browser against test servers in more than 300 cities, and your browser always talks to the closest one. The servers only send and receive random data. Your results are calculated on your device and are never uploaded.
After one warm-up request opens the connection, 20 empty requests measure the round trip. Server processing time reported by the test server is subtracted, the median becomes your ping, and the average change between samples becomes jitter.
Files are streamed for up to 11 seconds. Request size and the number of parallel streams grow with your speed, so slow and gigabit lines are both measured. The test stops early once the rate is stable.
Random, incompressible data is sent the same way, so compression on the path cannot inflate the result. Upload progress is sampled every 100 ms to draw the live graph.
While data is moving, ping is measured again. The increase over idle latency shows bufferbloat: queues in your router or ISP that cause lag in games and calls when the line is busy.
The final speed ignores the TCP ramp-up at the start, then averages one-second windows after dropping the slowest 30% and the fastest 10%. This gives a figure close to what large transfers achieve.
It depends on what you do and how many people share the connection. These are comfortable figures per activity. Add them up for simultaneous use.
| Activity | Download | Upload | Ping |
|---|---|---|---|
| Browsing, email, music | 5 Mbps | 1 Mbps | < 100 ms |
| HD video streaming (1080p) | 5 to 8 Mbps | Any | Any |
| 4K / UHD streaming | 25 Mbps | Any | Any |
| HD video calls | 4 Mbps | 4 Mbps | < 80 ms |
| Online gaming | 10 Mbps | 3 Mbps | < 40 ms, low jitter |
| Live streaming to Twitch / YouTube (1080p60) | Any | 8 to 10 Mbps | < 80 ms |
| Hosting a website or game server from home | Any | 10 to 50+ Mbps | < 60 ms |
| Large downloads and cloud backups | 100+ Mbps | 20+ Mbps | Any |
How fast data reaches you, in megabits per second (Mbps). It drives streaming quality, game and app downloads and page loads. Divide by 8 to get megabytes per second: 400 Mbps is about 50 MB/s.
How fast you can send data. It matters for video calls, live streaming, cloud backups and anything you host from home. Many cable and DSL plans are asymmetric, so upload is often 5 to 20 times lower than download.
The round-trip time to the test server while your line is quiet, in milliseconds. Lower is better: under 20 ms is excellent, 20 to 50 ms is good, and over 100 ms becomes noticeable in games and calls.
How much ping varies from one packet to the next. High jitter causes choppy voice, rubber-banding in games and video calls that freeze. Under 10 ms is great; above 30 ms usually means Wi‑Fi interference or congestion.
Ping measured while the download and upload are running. On a well-managed line it barely changes. A large increase means bufferbloat: lag spikes whenever someone else uploads a file or starts a stream.
A+ means latency rises by under 5 ms under load, A under 30 ms, B under 60 ms, C under 200 ms, D under 400 ms, and F beyond that. Routers with Smart Queue Management (SQM, fq_codel or CAKE) usually fix poor grades.
Plans are sold as “up to” a maximum. Many things between your device and the internet can lower it.
Yes. It is free with no account, no app and no ads. It runs in any modern browser on desktop and mobile.
The test runs on a global edge network with locations in more than 300 cities. Your browser is routed to the closest one automatically, and the panel shows which location served your test and roughly how far away it is.
No. Results are calculated in your browser and saved only in its local storage. The test servers only send and receive random data; your result is calculated on your device and never sent to them or to us. A shared link stores the numbers in the URL fragment, which is never sent to a server.
Roughly 10 MB on a slow connection and up to 1 GB on a gigabit line, because the test needs a few seconds of full-speed transfer in each direction. On a metered mobile plan, run it only when needed.
Every service measures to different servers over different routes and summarizes the data differently. Some report the peak, others an average. Differences of 10 to 20% are normal; large gaps usually point to a congested route to one provider.
Mbps (megabits per second) is how ISPs advertise speed. MB/s (megabytes per second) is what download managers display. One byte is eight bits, so 100 Mbps equals 12.5 MB/s. Use the unit switch on the test to view either.
Most cable, DSL and many fiber plans are asymmetric and reserve more capacity for downloads. If you host services, stream or back up to the cloud, look for a symmetric plan with matching upload and download.
Enable Smart Queue Management (SQM), QoS with fq_codel or CAKE, or an “adaptive QoS” feature on your router. Set it slightly below your measured speeds. This keeps latency low even when the connection is fully used.
Yes, if the upload speed is sufficient. Many ISPs use CGNAT or block inbound ports, which makes port forwarding impossible. A tunnel service such as Localtonet publishes local web apps, game servers and devices without port forwarding. Use the CGNAT check and port checker to diagnose your setup.
