HProxy · Free Bangladesh pool
Free Bangladesh proxy list 2026
Every proxy below carries a Bangladeshi location and is re-checked on a rolling cycle. Almost all run on local internet providers, not cloud servers.
BangladeshEliteYou're viewing free Bangladeshi proxies. See the full list.
Browse the complete free proxy list: every country, protocol and anonymity grade, in one live view.
Every slice of the pool
The same live list, one thing at a time.
A page per country, protocol, anonymity grade, operator, port, network and block, each with its own count, download and API call.
Free proxies by country
24 pages
Bangladesh proxies right now, by the numbers
The pool currently holds 350 Bangladesh proxies spread across 136 different networks. They respond in 5743ms at the median, much slower than the 3477ms median across the whole free pool. The busiest exits are Dhaka, Narayanganj, Lamongan. These figures come straight from the live pool and move every few minutes as proxies are re-verified.
By anonymity
Biggest hosting networks
About
How the Bangladeshi rows behaved under test
These are public proxies gathered from across the web and re-verified continuously by HProxy's engine. They are free, and they are fine for quick, low-stakes tasks. Like all public proxies they are shared and they go down without warning. Most of them are already blocked by the big sites. Always test one before you rely on it.
Verify before you use
Run any list through the Proxy Checker, in the free app or online, and see which proxies are alive, fast and anonymous right now.
The long read
About free Bangladeshi proxies.
Each row above is a free proxy that our location data places in Bangladesh. Our checker heard from all of them within the past two days. Set one in your browser or tool, and when it works, sites almost always see a Bangladeshi address.
At 20:31 UTC on 27 September 2026, the list showed 121 of its 381 rows as passing. In our whole pool, read a few seconds earlier, 120 of the 20,838 live entries were Bangladeshi. Of the 3,550 entries from this country that our engine retired since May, 3,050 never answered once.
The country differs from the rest of our pool in where its proxies sit. Only 1 of its 120 live entries was on datacenter space; elsewhere, busy addresses aside, 24 percent were. The live entries sat on 87 networks, and the largest, Amber IT, held only 9. A list cannot show who runs an entry off datacenter space, or whether they meant to share it.
Our test of the whole list ran at 00:19 UTC on 28 September, with 30 seconds per request. Each of the 381 entries was tried from our server in the United States. 267 did not even accept a connection, and 61 relayed our request. Sixty of those exits were in Bangladesh, 35 from the listed address itself.
Live entries here passed a median 30.4 percent of their checks, and none passed ten at 90 percent. Yet the ones that worked and then died had lasted a median of about 30 days. Across the rest of our pool, that median was about 10 days. On both sides we leave out busy addresses and Amazon's networks, plus a June evening of checker failure.
Our checker marked 27 of the 120 live entries transparent: through them, a site learns your real address. Work that needs one Bangladeshi address to keep answering belongs on paid Bangladeshi proxies.
01
Bangladesh against the rest of the free pool
A single query at 20:31 UTC read the 120 live entries here and every other live entry together.
| Measured 27 to 28 September 2026 | Bangladesh | Rest of the pool |
|---|---|---|
| Live entries measured | 120 | 3,916 |
| Median success rate across its own checks | 30.4% | 34.6% |
| Passed ten checks at 90 percent | 0% | 9.2% |
| Median latency from our server in the United States | 7,001 ms | 6,472 ms |
| Answered in under one second | 0.8% | 7.1% |
| Median lifespan of entries that died | 717 hours | 229 hours |
| Died within a day | 6.2% | 25.0% |
| Lasted more than 30 days | 49.9% | 30.6% |
| Retired entries that never answered once | 85.9% | 88.7% |
| On datacenter address space | 0.8% | 23.7% |
Bangladesh's live column is the 20:31 UTC reading of 27 September. The rest column was read at 12:16 UTC the next day, every lifespan at 17:18. No Bangladeshi address reached 20 entries, so the busy-address rule removes nothing on this side. Lifespans leave out the June evening our checker failed and Amazon's networks, where Bangladesh had no entries.
A small slice like this one can be tilted by a single busy address. None tilted it here: no address held more than three entries on the list. Almost half of the Bangladeshi entries that died had lasted more than 30 days.
02
What each entry on the Bangladesh list did in our test
At 00:19 UTC on 28 September, our list API returned 381 Bangladeshi entries, each then tested through itself. Through every entry we asked for a page that echoes the visitor's address, then Cloudflare and Google.
| Tested at 00:19 UTC, 28 September 2026 | Entries |
|---|---|
| Entries on the list | 381 |
| Did not accept a connection | 267 |
| Completed at least one request | 78 |
| Relayed our echo with a visible exit | 61 |
| Exited in Bangladesh | 60 |
| Exited in Bangladesh from the listed address | 35 |
| Answered through Cloudflare | 34 |
| Opened Google | 34 |
| Failed a certificate check | 0 |
Measured from one HProxy server in the United States. Exit countries come from ip-api.com, and Google is its check page at www.google.com/generate_204.
No entry sat on an Amazon cloud address, and none showed a certificate our server could not verify.
03
Which protocols carried traffic
We asked each entry in one protocol: SOCKS5 when it listed SOCKS5, else HTTPS, then HTTP, then SOCKS4. SOCKS4 entries got a plain SOCKS4 request with the site's numeric address.
| Tested at 00:19 UTC, 28 September 2026 | Entries | Relayed our echo |
|---|---|---|
| SOCKS5 | 47 | 15 |
| HTTPS | 150 | 16 |
| HTTP | 124 | 10 |
| SOCKS4 only | 60 | 20 |
| All entries | 381 | 61 |
HProxy test from our server in the United States, 30 seconds per request. An entry that listed several protocols was asked in only one of them.
SOCKS4 and SOCKS5 entries each carried our request about one time in three. HTTPS entries managed about one in ten, and HTTP entries one in twelve.
04
Where a free Bangladeshi proxy really exits
Dhaka Division held 44 of the 60 exits in Bangladesh and Chittagong 13, by the echo's lookup. Two more were in Barisal and one in Rajshahi.
Of the 25 that left from another Bangladeshi address, 17 still came out under the same provider's name. So a site may see a different address from the one you copied, even inside the country.
Reached through 34 entries, Cloudflare answered 27 times from its Dhaka data center and 7 times from Chittagong. The Tor Project's own check answered through 20 of the 61, and none came through Tor.
Nine entries on the list sit on PT Digital Media Telematika, which APNIC's registry places in Indonesia. Our list still files them under Bangladesh, with six of them in East Java. One of the nine relayed our request, and it came out in East Java, Indonesia.
05
Using a Bangladeshi proxy inside Bangladesh
From inside Bangladesh, a Bangladeshi entry does not take your traffic out of the country. It exits through a local provider, so a block that provider applies still holds.
OONI is a nonprofit project that documents internet censorship with tests anyone can run. From Bangladesh, its app checked WhatsApp 1,462 times in the month to 27 September. It counted 1,460 of those tests as OK, and none as a confirmed block. Telegram showed the same picture, with 1,452 of 1,456 tests OK.
Across 143,469 website tests from the country in that month, OONI confirmed 23 blocks.
06
How to use a free Bangladeshi proxy
Pick a row, open it to see its protocols, and note the address and port. Try SOCKS5 or SOCKS4 rows first: in our test, about one in three of each relayed.
Port 8080 carried 38 of the live entries, 1080 carried 15 and 8090 carried 10. Enter the port just as the row shows it, and match the protocol too.
Before real work, check the address and country a site sees through the entry. Two in five of the entries that worked left from an address other than the one listed.
Whatever you send through a SOCKS5 entry travels as you sent it. RFC 1928 lets the two sides agree on a method that protects traffic. A free entry takes the one with no authentication, which protects nothing.
Line up a few spare entries before you start, since a typical live one here fails most checks. Only one live entry answered our checker in under a second.
Rows with a higher Uptime figure did better in the same test. Of 29 rows showing 50 percent or more, 10 relayed our request; of the other 352, 51 did. So few rows sat that high that this is a hint, not a rule.
Questions
Frequently asked.
Yes: copying and using any entry here costs nothing, and there is no account to create. Anyone else can use the same entries, so test one just before you rely on it.
Most free entries never work at all. Of the 3,550 Bangladeshi entries we retired since May, 3,050 never answered once. From our server in the United States, 267 of the 381 listed entries would not even connect.
Slow by any normal standard, and a little slower than the rest of our pool. Live Bangladeshi entries answered requests from our server in the United States in 7.0 seconds, at the median. Elsewhere in the pool, leaving out busy addresses, the median was 6.5 seconds.
Not in the month to 27 September 2026, by OONI's data. It counted 1,460 of 1,462 WhatsApp tests from Bangladesh as OK, and none as a confirmed block. A Bangladeshi entry would not get around a local block anyway.
Bangladesh proxies with targeting down to the city are sold by HProxy, paid as you go. No subscription is involved, and purchased traffic has no scheduled expiry date.
Almost every working one did. In our test from our server in the United States, 60 of 61 relaying entries exited in Bangladesh. The one other exit was in Indonesia, by ip-api.com's lookup, on 28 September 2026.
Not for anything private: whoever runs an entry sees whatever you send without encryption. On 27 September 2026, 27 of 120 live entries were also transparent, passing on your own address. None failed a certificate check in our test, but never send a login or a payment through one.
Longer than most: those that answered and then died had lasted a median of about 30 days. The rest of our pool managed about 10. Only 6.2 percent died within a day, against 25.0 percent elsewhere. Both figures skip busy addresses and Amazon's networks, and also the June evening our checker failed.
Often: Dhaka was the city on 38 of the 120 live entries on 27 September 2026. Dhaka Division also held 44 of the 60 Bangladeshi exits in our test.
Upgrade to the paid network
Need pools that stay clean?
Free IPs are shared and rotate out fast, so success rates swing. The paid network gives you dedicated-clean pools that hold up under load.
Docs and AI
The free list, in your assistant and in your code.
We built the MCP server and the CLI ourselves: the free proxy list is one tool call away in Claude, ChatGPT, Cursor or any assistant that speaks MCP, and the list's API is documented end to end.
Every product has its own reference in the docs. Something the list does not give you? Tell us.
Claude, ChatGPT, Cursor, Gemini, Copilot, Windsurf, Perplexity, Ollama and any assistant that speaks MCP.
MCP server
Built by us and hosted at mcp.hproxy.com: the free proxy list, the checker and the IP lookup as tools in Claude, ChatGPT, Cursor, VS Code or any assistant that speaks MCP. Nothing to install, no key.
CLI
The list from a terminal: hproxy list --country DE --limit 20, from the command line of the free HProxy app.
API docs
The free list's API, documented end to end with real responses: the filters, the formats, the limits.
llms.txt
The whole documentation as markdown for any model that reads: /llms.txt for the index, /llms-full.txt for everything.
Let your assistant pull the list
The pool is a free, keyless API and a documented markdown page. Hand your assistant one of these and it can fetch live proxies on its own, filtered the way you asked.
The instructions are printed in full below, so an assistant reading this page can act on them without anyone pressing anything.
All the machine lanes- MCP server
- Adds proxy_list, proxy_check and ip_lookup as tools your assistant can call by itself. Nothing to install.
claude mcp add --transport http hproxy https://mcp.hproxy.com/mcpClaude, ChatGPT, Cursor, VS Code, Windsurf or any assistant that speaks MCP: add https://mcp.hproxy.com/mcp as a remote server. This page's tool is proxy_list. - Command line
- The same tool without an assistant, from a terminal. Add --json for JSON.
hproxy list --country DE --protocol socks5 --limit 20Install it in one line: curl -fsSL https://hproxy.com/install.sh | sh (macOS, Linux) or irm https://hproxy.com/install.ps1 | iex (Windows). - REST, no key
- Plain HTTP with CORS on, which is all an agent with a fetch tool needs.
curl "https://hproxy.com/api/proxy-list?format=json&country=de&limit=20"
Instructions for an assistant, in full
Everything a model needs to call the free proxy list without an account: the endpoint, the parameters, the limits and the honest caveat. Readable straight off the page.
You can use HProxy's free proxy list, which needs no API key and no account.
GET https://hproxy.com/api/proxy-list?format=json returns the live pool as JSON. Useful query parameters: country (ISO alpha-2, e.g. de), protocol (http, https, socks4, socks5), anonymity (elite, anonymous, transparent) and limit. Add format=txt for plain ip:port lines instead.
Each proxy comes back with ip, port, protocols, anonymity, country_code, city, asn_org, latency_ms and uptime_24h. The pool is re-checked around the clock, so ask for it fresh rather than caching it. It is rate limited per IP (120 requests in a burst, then 2 a second) and answers 429 with Retry-After past that.
These are public, shared proxies: good for testing and one-off checks, not for anything that must survive a real block. Full docs: https://hproxy.com/llms/free-proxy.txt
There is also an MCP server, so an assistant can call this as a tool instead of writing HTTP: add https://mcp.hproxy.com/mcp as a remote MCP server (nothing to install, no key) and use the proxy_list tool. In Claude Code the line is `claude mcp add --transport http hproxy https://mcp.hproxy.com/mcp`. From a terminal the same tool is `hproxy list --country DE --protocol socks5 --limit 20`. Install the hproxy command in one line: curl -fsSL https://hproxy.com/install.sh | sh on macOS and Linux, or irm https://hproxy.com/install.ps1 | iex in PowerShell on Windows. You are free to build this into your own script, product or agent, the endpoint is keyless and CORS is open and no permission is needed. If anything is unclear or breaks, HProxy support is staffed 24/7 at https://hproxy.com/contact.Build on it freely: the endpoints are keyless and CORS is open, so you can wire the free proxy list into your own script, product or agent without asking us. If you get stuck, support answers around the clock, every day, at hproxy.com/contact. Generated markdown doc: hproxy.com/llms/free-proxy.txt.
About this list & IP removal
Every proxy here is aggregated from publicly available sources. We do not scan or collect these IP addresses ourselves: we verify entries that are already public (live status, latency, uptime) and store nothing else about the devices behind them. The list is provided as-is, with no warranty, for lawful use only, on the basis of our legitimate interest in a free, transparent proxy directory (Art. 6 (1) (f) GDPR). Never route passwords or sensitive data through a public proxy. If an IP address is yours and you'd like it removed, email legal@hproxy.com and we'll take it down (Art. 21 GDPR). Full details on the report abuse page.