Asia availability

Free Hong Kong Proxy List

Live free Hong Kong proxies: HTTP, HTTPS, SOCKS4 & SOCKS5, re-verified every 5 minutes with uptime and speed checks. Free TXT/CSV/JSON + API.

live endpoints
60live endpoints
verification window
Liveverification window
refresh cadence
5 minrefresh cadence
Hong Kong live evidenceAsia

Observed routes

Hong Kong44
Chengdu1
Los Angeles1

60live nodes

1500 msmedian latency

65%elite-grade

65%Google-pass

Observed cities: Hong Kong · Chengdu · Los Angeles

A live Databay API snapshot captured 2026-07-19T13:36:13.201Z returned 51 unique endpoints labelled for Hong Kong. The retrieved Hong Kong rows were 32 HTTP (63%), 17 SOCKS5 (33%), 2 SOCKS4 (4%). Hong Kong checker header classifications were 26 ELITE (51%), 21 TRANSPARENT (41%), 4 ANONYMOUS (8%), with a median checker latency of 1,859 ms and a mean reported uptime of 81.9%. The newest Hong Kong row-level check in this capture was 2026-07-19T13:32:34Z; the oldest was 2026-07-19T08:06:28Z. These Hong Kong observations come from a volatile public-relay list and do not promise that the same endpoints will remain reachable. In the Hong Kong rows, “elite”, “anonymous”, SSL, Google, latency, and uptime are fields reported by the checker for particular probes; they do not prove operator identity, permission, security, residential origin, city accuracy, or suitability for sensitive traffic. Re-query the live Hong Kong endpoint before every authorized test, validate the exit independently, and never send credentials, personal data, production cookies, or confidential payloads through an unknown relay.Hong Kong should be tested as its own locale and commercial context, with Traditional Chinese or English, Hong Kong-dollar display, delivery, account, and device state recorded. Do not treat a Hong Kong exit as interchangeable with mainland China. Use a Hong Kong-geolocated public proxy only as one disposable country-level network-origin sample for an authorized localization, catalog, advertising, search, accessibility, or delivery-message check. For the Hong Kong comparison, hold the destination URL, timestamp, account, cookies, browser, language, currency, address, and experiment steps constant, then compare with a direct or controlled reference. Stop the Hong Kong test when the source blocks, challenges, rate-limits, or disallows the workflow. Confirm any consequential Hong Kong finding through an official API, licensed feed, controlled paid exit, local reviewer, or the source itself before making a product, compliance, pricing, or enforcement decision.

Live proxy list

Free Hong Kong proxies

Re-verified every 5 minutes

Showing 110 of 60

This table scrolls horizontally when all columns do not fit.

IP:PortProtocol
hkHTTPS1096 ms
hkSOCKS5967 ms
hkSOCKS5965 ms
hkSOCKS51429 ms
hkHTTP458 ms
hkHTTPS804 ms
hkSOCKS51197 ms
hkHTTPS2813 ms
hkSOCKS51417 ms
hkHTTP425 ms

Keep going

Need filtered proxy downloads?

This list is free to use. When a job needs a managed uptime target and support, compare Databay’s paid pools.

Free · this page

Grab the list, no account

  • Full TXT/CSV/JSON download
  • Verified every 5 minutes
  • Search, sort & copy any proxy
  • Free API, 60 requests/min
Premium · production

Filtered exports & managed uptime targets

  • Filter by protocol, country & anonymity
  • Residential & datacenter pools, 99.9%+ uptime
  • API keys, higher rate limits, support

Standard residential self-serve orders start at 1 GB.

FAQ

Frequently Asked Questions

Are these free proxies safe to use?
Do not assume a public proxy is safe. Its operator may inspect or log traffic, and a successful Databay check is not a security review. Use the list only for non-sensitive testing, never for credentials, personal data, payments, or private accounts.
What is the difference between a free proxy list and a free web proxy?
A proxy list provides IP addresses and ports that you configure. A web proxy is a website where you paste a URL. Databay provides a server list, not a web proxy.
How often is the list updated?
Candidates are checked on a rolling five-minute schedule. Use Last Checked for a specific result. The primary pool uses checks from the past six hours; a 24-hour fallback applies only when fewer than 50 endpoints qualify.
What is the difference between HTTP, HTTPS, SOCKS4 and SOCKS5 proxies?
HTTP forwards plain web requests, while HTTPS-capable HTTP proxies relay TLS with CONNECT. SOCKS4 relays TCP. SOCKS5 can add authentication, remote DNS, and optional UDP support.
What do elite, anonymous and transparent mean?
These are header observations, not security ratings. Elite exposed neither the tested client IP nor common proxy markers; anonymous exposed a proxy marker but not the tested IP; transparent exposed the tested IP.
How do I get a free proxy from this list?
Choose a recently checked row, copy its IP address and port, and select the matching protocol in your client. For bulk use, download TXT, CSV or JSON, or call the free API. No signup is required.
Is there a free API?
Yes. The /api/v1/proxy-list endpoint returns the live list as JSON, CSV or TXT with filters for protocol, country, SSL, Google-pass, speed and anonymity. No key required.
Can I filter by country or protocol?
Yes. Use the country pages, the protocol pages (HTTP, HTTPS, SOCKS4, SOCKS5) and the anonymity pages (elite, anonymous, transparent), or the API query parameters.
Why do some proxies stop working quickly?
Public endpoints can become overloaded, change configuration, or go offline without notice. A result only describes the latest successful check, so expect failures and use timeouts and retries for testing.
Do these proxies support HTTPS?
Some do. Strict means the checker completed an HTTPS request with certificate validation; Loose required validation to be disabled; None means no HTTPS tunnel completed. Do not send sensitive data through a public operator.
What does "Google passed" mean?
It records that the upstream checker received its expected Google probe response for that route and time. It is not a reputation, safety, or future-access guarantee.
Should I use free proxies for scraping or automation?
Use public proxies only for permitted, non-sensitive experiments. They are too unstable for dependable automation and do not override terms, robots controls, authentication, or rate limits.
What did the Hong Kong proxy API return in this snapshot?
The Hong Kong capture retrieved 51 unique IP:port endpoints from the country-filtered Databay API. Hong Kong protocol labels were 32 HTTP (63%), 17 SOCKS5 (33%), 2 SOCKS4 (4%); header classifications were 26 ELITE (51%), 21 TRANSPARENT (41%), 4 ANONYMOUS (8%). The Hong Kong sample had a median checker latency of 1,859 ms and a mean reported uptime of 81.9%. These Hong Kong figures are a dated observation, not a live inventory guarantee; use https://databay.com/api/v1/proxy-list?country=HK for the current result.
Does a Hong Kong proxy prove a local user or precise location?
No. The Hong Kong label is a geolocation observation about an exit address and may be stale or wrong. A Hong Kong label does not establish a household, operator, consent, city, language, account, delivery address, tax state, device, or eligibility. For important Hong Kong work, verify the exit and use controlled infrastructure plus a local or primary-source confirmation.

Need managed proxies for production?

Free lists can help with low-stakes testing. For production workloads, Databay’s managed pools publish a 99.9%+ uptime target.

Paid residential plans are separate from the free proxy list.