Free Tool
Proxy Speed & Availability Checker
Benchmark any proxy — residential, mobile 4G/5G, static ISP, datacenter, SOCKS5 or HTTP/HTTPS — against real global endpoints. Tune the samples, timeout and concurrency, then save each run so you can compare providers side by side.
Works with every proxy type
Residential, mobile 4G/5G, static ISP, datacenter and IPv6 endpoints are all supported. Every input is saved to the URL — copy it to share the exact result.
Test settings
Configure your proxy in this browser first, then run the test. Every request is made from your machine — results reflect your real exit-node performance.
Endpoints
Saved results
No saved runs yet. Run a test and click Save result — runs are stored in this browser only, never uploaded, so you can benchmark one provider, switch proxy, and compare the numbers directly.
Reading the numbers
Timings include DNS, TCP, TLS and — when a proxy is active — the SOCKS5 or CONNECT negotiation, so they will always be higher than a raw ICMP ping. Availability below 98% on a paid residential pool is a red flag; jitter above 150 ms usually means an overloaded gateway or a mobile exit sharing a congested cell.
Note: cross-origin responses are fetched in no-cors mode, so throughput is estimated from the known payload size of the Cloudflare test endpoint rather than measured bytes. Run the test three times and take the median before drawing conclusions.
01 · Overview
What is the Proxy Speed & Availability Checker?
The Proxy Speed & Availability Checker turns the vague question 'is this proxy fast?' into four hard numbers: availability (what percentage of requests actually complete), average latency, jitter (how consistent the gateway is) and estimated throughput. It runs entirely in your browser through whichever proxy your browser or operating system is configured with — residential, mobile 4G/5G, static ISP, datacenter, SOCKS5 or HTTP/HTTPS — and hits real-world endpoints in North America, Europe, Asia and Oceania rather than a synthetic speed server. Every test setting is adjustable, and every run can be saved locally so you can benchmark provider A, switch credentials, benchmark provider B and compare the two with identical parameters.
02 · How it works
How the tool works
- 01Configure your proxy in the same browser that has this page open, then reload so the tunnel is active.
- 02Choose your test settings: samples per endpoint, request timeout, concurrency, and whether to discard a warm-up request.
- 03Select which global endpoints to hit, or paste a custom URL to benchmark the exact target you scrape.
- 04Run the test — each endpoint is timed sample by sample and the table fills in live with min, average, max, jitter and success rate.
- 05Save the run with a descriptive name. It is stored in your browser only, and appears under Saved results for later comparison or JSON export.
03 · Use cases
When to use this tool
Compare two providers fairly
Identical endpoints, samples and timeout for both runs — the only variable left is the proxy itself.
Validate an SLA before renewal
Run 20 samples per endpoint across a week and check whether availability really holds at the 99.9% the provider advertises.
Pick a region for a scraping job
Latency per endpoint shows which continent your current exit pool actually serves well.
Diagnose flaky sessions
High timeout counts with low latency on successful requests points at gateway throttling, not distance.
Benchmark your real target
The custom endpoint field lets you time the exact site you scrape instead of a generic CDN.
04 · Pro tips
Best practices
- →Always run a baseline test with the proxy disabled first — the difference is your true proxy overhead.
- →Use at least 5 samples; a single request is dominated by TLS handshake noise.
- →Keep concurrency at 1 when measuring latency and raise it to 4–8 when stress-testing gateway capacity.
- →Repeat the test at peak hours in your target market — residential pools degrade when their peers are online.
- →Save every run with the provider, plan and exit country in the name so the history stays readable.
05 · FAQ
Frequently asked questions
How does the proxy speed and availability checker work?+
The tool issues timed HTTPS requests from your own browser to a set of global endpoints (Cloudflare, Google, Amazon, BBC, Baidu, ABC Australia, Instagram and a 100 KB Cloudflare download). Whatever proxy your browser or OS is configured with carries those requests, so the latency, jitter, success rate and throughput you see are your real exit-node numbers. Nothing is proxied through 5Proxy servers and no results leave your machine.
What test settings can I change?+
Samples per endpoint (1–20), request timeout (2–30 seconds), concurrency (1–8 parallel requests), an optional warm-up request that is discarded to exclude TLS handshake cost, which endpoints to include, and a custom endpoint URL so you can benchmark the exact target you scrape.
Where are my saved results stored?+
Saved runs live in your browser's localStorage under a single key — they are never uploaded, never tied to an account, and never visible to 5Proxy. You can keep up to 25 runs, load any of them back into the results table, export the whole history as JSON, or clear everything with one click.
What is a good availability score for a residential proxy?+
98–100% is what a paid residential or ISP pool should deliver. 95–98% is acceptable for rotating mobile 4G/5G exits, where cell hand-offs cause occasional drops. Anything below 95% on a paid plan means the gateway is overloaded or the pool is recycling dead peers — raise it with support or test another provider.
Why is my proxy latency higher than a ping test?+
This test measures full application round-trip time: DNS resolution, TCP handshake, TLS negotiation, the SOCKS5 or HTTP CONNECT negotiation with your proxy, and the response. A ping only measures ICMP round-trip to the first hop. Expect 3–6x the ping value on a residential exit and more on mobile.
How do I compare two proxy providers with this tool?+
Run the test with provider A configured, save the run with the provider's name, switch your browser to provider B, run the identical settings again, and save. Both runs appear in Saved results with availability, average latency, jitter and peak throughput side by side, so the comparison uses the same endpoints, samples and timeout.
Does high jitter matter for scraping?+
Yes. Jitter above roughly 150 ms means inconsistent gateway performance, which forces you to set long client timeouts and lowers effective concurrency. For checkout automation, ad verification and sneaker copping, low jitter matters more than low average latency.
Is the throughput number accurate?+
It is an estimate. Browsers block reading cross-origin response bodies, so the tool derives Mbps from the known payload size of the Cloudflare 100 KB test endpoint divided by the measured round trip. Treat it as a relative indicator when comparing two proxies with identical settings, not as an absolute bandwidth figure.
Can I test SOCKS5, HTTP and mobile proxies with it?+
Yes. The checker is protocol-agnostic — it tests whatever path your browser is using, so SOCKS5, SOCKS5h, SOCKS4a, HTTP, HTTPS, IPv4 and IPv6 endpoints all work, across residential, mobile 4G/5G, static ISP and datacenter proxies.
07 · Related tools
Continue your audit
Free tool
Global Latency Test
Measure round-trip latency from your current connection to popular scraping targets across five global regions.
Free tool
My IP & Location
Instantly reveal the public IP, ASN, ISP, country, city and proxy/VPN flags your current connection exposes.
Free tool
Bandwidth Calculator
Estimate monthly SOCKS5 bandwidth spend from request volume, average page size, overhead and per-GB pricing.
Also known as