FindMyIP is a free, all-in-one diagnostic tool for checking your public IP address, testing your internet speed, and verifying which ports are open on your connection. Everything runs in your browser — no signup, no tracking, and your test history stays local on your device. Whether you are troubleshooting a slow Wi-Fi connection at home, verifying a port forwarding rule for your game server, or documenting consistent network problems for an ISP complaint, FindMyIP gives you the data you need in one place.
The tool is built for IT professionals, gamers, self-hosters, remote workers, and anyone diagnosing a network problem. It combines what most people open three separate tabs for: an IP address lookup, an internet speed test, and a real-time port checker. Speed measurements use Cloudflare's global edge network with peak 1-second window analysis to closely match the readings you would get from professional native applications like Speedtest. Port checks run through a dedicated backend that performs real TCP connections from outside your network, giving you accurate, real-time port reachability data rather than the cached scan results most free port checkers rely on.
When you run a speed test, the application opens multiple parallel connections to a test server and either downloads or uploads as much data as possible within a fixed time window — usually ten seconds. The total bytes transferred divided by the elapsed time gives you the throughput in megabits per second. The reason professional speed tests use parallel connections is that a single TCP stream cannot saturate a fast internet connection due to congestion control algorithms. With four to eight streams running simultaneously, the test can fully utilize the available bandwidth and produce numbers that reflect what your line can actually do.
One detail that separates accurate speed tests from inaccurate ones is how they handle TCP slow-start. When a new connection opens, TCP intentionally starts slowly and ramps up to find the maximum sustainable rate. This ramp-up phase can take one to two seconds on a fast connection. Tests that average throughput across the entire ten-second window pull the result down by including this slow ramp-up period. FindMyIP discards the first two seconds of every test and reports the peak sustained throughput over any one-second window during the remaining time — the same methodology used by leading speed test services.
Upload measurements work the same way in reverse: the browser pushes randomly-generated payloads to the server, counts bytes acknowledged over time, and trims the slow-start period. Because some networks compress traffic on the wire, FindMyIP uses cryptographically random bytes that cannot be compressed, ensuring the measured throughput reflects raw line capacity rather than compression efficiency.
Latency, also called ping, is the time it takes for a single packet to travel from your device to a server and back, measured in milliseconds. Anything under 30 milliseconds feels instant for typical web browsing. Below 50 milliseconds is generally considered excellent for online gaming and video calls. Between 50 and 100 milliseconds is acceptable but you may notice slight delays. Above 100 milliseconds the experience degrades noticeably for real-time applications.
Jitter is the variability in latency measurements over time. A connection might have a great average latency but if the jitter is high, individual packets arrive unpredictably late, which manifests as choppy video, dropped audio in calls, and rubberbanding in games. A jitter value under 10 milliseconds is excellent, 10 to 30 milliseconds is acceptable for most uses, and above 30 milliseconds typically indicates a network quality problem worth investigating.
Bufferbloat is one of the most important and least understood network problems. It happens when your router or modem queues up too many packets when the connection is busy, causing latency to spike dramatically under load. Your idle latency might be a healthy 10 milliseconds, but if downloading a file pushes it to 200 milliseconds, that is bufferbloat. The result is video calls breaking up exactly when other people in your house start streaming, gaming lag spikes whenever a Windows update kicks off, and websites that load slowly during busy hours. FindMyIP measures download and upload latency separately so you can spot bufferbloat at a glance: if your download or upload latency is much higher than your idle latency, your router is likely the culprit. Many modern routers have a setting called SQM (Smart Queue Management) or QoS that can reduce or eliminate bufferbloat.
A port is a numbered endpoint on your network where specific types of traffic arrive and leave. There are 65,535 possible ports, divided into well-known ports (1 to 1023 for standard services), registered ports (1024 to 49151 for applications), and dynamic ports (49152 to 65535 for temporary use). When you visit a website, your browser connects to port 80 for HTTP or port 443 for HTTPS on the web server. When you check email, your client typically uses port 587 to send and port 993 to receive. Each application listens on its own port.
For most home users, all incoming ports on their public IP are closed by default — the router blocks unsolicited traffic from the internet. This is a good thing for security. However, if you run a service that needs to accept incoming connections — a Minecraft server, a self-hosted media library, a security camera you want to view remotely, a VPN, or a remote desktop connection — you need to configure port forwarding on your router and confirm the port is actually reachable from the outside world. This is where a port checker becomes essential.
FindMyIP performs a real TCP connection attempt from a server outside your network to the port number you specify on your public IP. If the connection succeeds, the port shows as Open. If the connection is actively refused, it shows as Closed (meaning your router received the request but nothing was listening). If the request times out with no response, it shows as Filtered (meaning a firewall is silently dropping packets). These distinctions matter when troubleshooting: an Open port confirms your forwarding works, a Closed port suggests the forwarding is set up but no application is running on it, and a Filtered port means a firewall — possibly your ISP's, possibly your own — is blocking the traffic before it reaches your network.
Here are the most commonly checked ports and what they are typically used for, useful when troubleshooting which services need to be reachable on your network:
Port 21 (FTP) — File Transfer Protocol, used for transferring files between computers. Largely replaced by SFTP and HTTPS for security reasons.
Port 22 (SSH) — Secure Shell for remote command-line access to Linux and Unix servers. Also used for secure file transfer via SFTP.
Port 23 (Telnet) — Unencrypted remote terminal access. Considered obsolete and should not be exposed to the internet under any circumstances.
Port 25 (SMTP) — Simple Mail Transfer Protocol for sending email. Most ISPs block outbound port 25 to prevent spam.
Port 53 (DNS) — Domain Name System. Translates domain names like example.com into IP addresses.
Port 80 (HTTP) — Standard web traffic, unencrypted. Most web servers also listen on this port to redirect to HTTPS.
Port 110 (POP3) — Post Office Protocol for downloading email. Largely replaced by IMAP.
Port 143 (IMAP) — Internet Message Access Protocol for reading email from a server.
Port 443 (HTTPS) — Encrypted web traffic. The standard port for any secure website.
Port 3389 (RDP) — Remote Desktop Protocol for Windows remote desktop access. Should be protected behind a VPN if exposed.
Port 8080 (HTTP-Alt) — Alternative HTTP port, commonly used by web applications and proxies.
Internet service providers sell connection plans based on advertised speeds. If you pay for a 500 megabit per second plan but consistently see 200 megabits per second over weeks of testing, you have a legitimate basis for a complaint. The key word is consistently — a single bad reading on a busy evening is normal, but a pattern of underperformance across many tests at different times of day is a real problem.
FindMyIP supports this kind of long-term documentation. The auto-test feature runs ten tests at scheduled intervals and stores the history locally in your browser. The export functions create a CSV file you can analyze in a spreadsheet, or a printable HTML report with summary statistics and a complete test log. Both of these are useful when escalating a complaint: most ISP customer service representatives are more responsive when presented with concrete data rather than vague complaints, and regulators like the Federal Communications Commission in the United States or Ofcom in the United Kingdom will ask for measurement logs when investigating complaints.
When preparing a complaint, focus on numbers that clearly show underperformance: the average download speed across many tests, the lowest speed measured, the number of tests that fell below a critical threshold, and the latency under load. A line that delivers 200 megabits per second on average when sold as 500 is a problem. A line that delivers 480 megabits per second on average with occasional spikes to 510 and dips to 450 is delivering what was promised, even if the marketing implied a guaranteed minimum.
For download and upload speeds, results typically land within approximately 10 percent of Speedtest.net on the same connection. FindMyIP uses 8 parallel TCP streams against Cloudflare's edge network, trims TCP slow-start ramp-up, and reports the peak sustained throughput over a 1-second window — the same technique used by professional speed test tools. Latency tends to read 5 to 15 milliseconds higher than native apps because browsers cannot send ICMP packets and have to use HTTP requests, which include unavoidable server processing overhead. This is a limitation of every browser-based speed tester, not just FindMyIP.
Yes. Unlike sites that rely on cached data from services like Shodan, FindMyIP uses a dedicated backend that performs a live TCP connection to your public IP from outside your network. If a port shows as Open, traffic from the internet really can reach that port right now. If it shows as Closed, the connection was actively refused. Results are real-time, not cached. The backend enforces same-IP scanning only, meaning you can only check your own public IP, which prevents abuse of the service for scanning other networks.
All test history is stored only in your browser's local storage. We do not run analytics, do not have a database tracking users, and do not share your IP address with anyone beyond the third-party APIs needed to look up your ISP information. The third-party services we use are ipinfo.io for IP lookup and Cloudflare for the speed test itself. The port checker enforces same-IP-only scanning, meaning you can only scan your own public IP, not anyone else's. See our Privacy Policy for the complete details.
Browser-based ping measurements use HTTP requests, which add 5 to 15 milliseconds of fixed overhead from TLS handshakes and server processing time on top of the actual network round trip. The native ping command uses ICMP packets that bypass all of that overhead. Our tool calibrates and subtracts most of this overhead automatically, but a small gap from native ping is unavoidable — and applies to every browser-based speed tester, including Speedtest's web version. If you need a true ICMP ping measurement, use the ping command in your operating system's terminal.
Idle latency is the round trip time when your connection is not being used heavily. Download latency is measured during a download, and upload latency during an upload. The gap between idle and loaded latency is called bufferbloat. A high gap, typically 60 milliseconds or more, indicates your router or modem is queueing packets aggressively, which can hurt video calls and gaming even if your raw speed numbers look fine. If you see significant bufferbloat, check whether your router supports SQM or Smart Queue Management, which can largely eliminate the problem.
Yes. After running a few tests, click Export CSV for a spreadsheet of all your readings, or Generate Report for a printable HTML document with summary statistics and a full test log. Both are useful for filing complaints or proving your line is underperforming. ISP customer service representatives tend to be more responsive when presented with concrete data showing consistent underperformance across many test runs at different times of day.
This is normal and reflects real differences between devices. WiFi link quality is per-device, so signal strength and modulation differ even on the same network. CPU performance matters at high speeds because speed tests are partly limited by how fast your device can decrypt TLS-encrypted data. Background applications using bandwidth also affect results. Browser HTTP stack implementations differ between Safari, Chrome, and Firefox in ways that can affect measurement accuracy. All these factors combine to mean a 10 to 20 percent difference between devices on the same network is expected.
Each speed test transfers approximately 250 to 500 megabytes total, depending on your connection speed. The auto-test feature is capped at 10 runs maximum to prevent accidentally consuming large amounts of bandwidth if you forget about it. After 10 runs, the auto-test automatically stops and resets. You can manually stop it at any time by clicking the Stop button.