The useful question is never whether fyvri/fresh-proxy-list is good. It is whether it fits the job you are about to run, on your target, this month. A free proxy list that suits one workload is wrong for the next one.
Everything we can stand behind
fyvri/fresh-proxy-list trades at github.com/fyvri/fresh-proxy-list. They compete with us as a free proxy list.
There is no price in this comparison because we do not have one we can date and attribute. What follows instead is the set of questions that decide the cost once you do read their page.
The four figures worth copying down are the rate without a commitment, the smallest purchase allowed, what happens to unspent balance, and whether a failed request still costs money.
fyvri/fresh-proxy-list in its own words
The title tag on 9 September 2026 was "GitHub - fyvri/fresh-proxy-list: An automatically updated list of free HTTP, HTTPS, SOCKS4, and SOCKS5 proxies, available in multiple formats including JSON, TX". Their meta description, the sentence Google shows under the link, says: "An automatically updated list of free HTTP, HTTPS, SOCKS4, and SOCKS5 proxies, available in multiple formats including JSON, TXT, CSV, XML, and YAML. - fyvri/fresh-proxy-list"
Taken from the same page, on the same day:
The product lines named on that page are SOCKS5. Support is advertised as around the clock. Their pricing page is titled "Pricing · Plans for every developer · GitHub".
All of the above is their copy, not our finding. A dated quotation is the weakest kind of evidence that is still evidence.
How far back the record goes
The domain was registered on 9 October 2007, through MarkMonitor Inc, which makes it about 18 years old. Checked against the registry on 9 September 2026. The registration currently runs to 9 October 2028.
None of it predicts success rate. It is here because domain age and archive depth are free to check, and a provider whose public record starts last quarter deserves a smaller first order.
The questions behind the numbers
Shared or dedicated changes everything
A shared address arrives with a history you did not create and cannot see. For read-only work it rarely matters. For anything that keeps a session, it is the most common cause of unexplained failure.
The country total, and the country you use
Ask for the pool size in your country rather than in total, and expect the answer to be vague. Where it is not published, the ten minute test at the end of this page settles it for a few dollars.
The response time nobody advertises
Every provider works when nothing is wrong. The difference shows up on the day a subnet gets flagged or a gateway stops answering. Before committing volume, send one real question through the support channel and time the reply.
The comparison, honestly framed
Compare on these rows and ignore the feature lists on both sites:
| Question to ask | fyvri/fresh-proxy-list | HProxy |
|---|---|---|
| Entry price | Not verified by us, read their page | From $0.44/GB, pay as you go |
| Does unused balance expire? | Not published clearly, ask before paying | Never |
| Minimum commitment | Check the current terms, this one moves | None |
| Identity checks | Their terms page answers this | Card orders are screened, and verification is sometimes requested |
| Targeting | Worth confirming in writing before you commit | Country, city and ISP |
| Free way to evaluate | Read their current terms | Free proxy list and proxy checker |
Treat the right-hand column as a claim by an interested party, because it is one. Every line in it is checkable on our own pages and in our own terms.
The test that beats this page
It is a small amount of money for an answer that is actually about your target.
- Take the smallest paid unit each provider will sell you. A free trial works for a first pass, though trials sometimes ride better pools than paid plans do.
- Fire the same batch through both on one afternoon. Different days are different conditions, and that ruins the comparison.
- Write down two numbers per provider: the share of requests that succeeded, and the traffic it cost.
- Take a handful of the addresses you were given and verify their country and type against a checker you do not control.
- Choose on the measurement, not the marketing, and expect the answer to change within a year.
Comparable providers
We hold the same record for these, and the same honest limits apply to each:
- PremProxy alternatives
- Proxifly (GitHub) alternatives
- Proxy Daily alternatives
- Proxy IP List alternatives
- Proxy List alternatives
- Proxy-Listen.de alternatives
Where we fit
For completeness: residential from $0.44 per GB, city and ISP targeting included, no monthly minimum, and a balance that stays where you left it. We are a good fit for uneven workloads and a poor one for anybody who wants a signed enterprise agreement.
