Benchmark report
Is Wikipedia slow?
Portal load times from 6 cities
Slow from some locations
Tokyo waited 427 ms, about 31× Amsterdam. The gap suggests some test locations reach a distant server.
- Fastest
- 14 ms
- Amsterdam
- Slowest
- 427 ms
- Tokyo
- Successful requests
- 100%
- 6 locations
Response time around the world
Amsterdam
14 msSingapore
16 msMontreal
84 msMumbai
324 msSan Francisco
388 msTokyo
427 ms
Why are San Francisco, Tokyo and Mumbai slower?
In the controlled run of 4 October Amsterdam and Singapore finished in 19 and 20 ms, connecting in 6 and 3 ms. Montreal took 80 ms: 15 ms to connect, 19 ms for TLS, 19 ms waiting and 27 ms downloading. San Francisco and Tokyo took 384 and 422 ms and Mumbai 321 ms, and their first phase gives the reason: 58 to 79 ms just to open the connection, then 64 to 81 ms for TLS, 74 to 99 ms waiting and 125 to 163 ms downloading the 120 KB over that long path. Tokyo and Mumbai connected to the address Singapore used (103.102.166.224), Wikimedia's nearest caching site to either city, so their numbers are distance. San Francisco connected to the address Montreal used (208.80.154.224). Wikimedia operates a caching site in San Francisco; our worker there does not appear to reach it.
Click the location to see each stage.
Tokyo422 ms
- Finding the server
- 1 ms
- Reaching the server
- 79 ms
- Setting up security
- 81 ms
- Waiting for the server
- 99 ms
- Receiving the response: most of the time
- 163 ms
- Total
- 422 ms
Finding the server is measured once per location, so it sits outside these bars. Open a city to see it.
Bars show the typical request, so the parts add up to its total. Why
Compare all locationsHide the comparison
Click a city to see its stage-by-stage breakdown.
Amsterdam19 ms
- Finding the server
- 2 ms
- Reaching the server
- 6 ms
- Setting up security: most of the time
- 8 ms
- Waiting for the server
- 3 ms
- Receiving the response
- 2 ms
- Total
- 19 ms
San Francisco384 ms
- Finding the server
- 13 ms
- Reaching the server
- 70 ms
- Setting up security
- 76 ms
- Waiting for the server
- 89 ms
- Receiving the response: most of the time
- 149 ms
- Total
- 384 ms
Montreal80 ms
- Finding the server
- 2 ms
- Reaching the server
- 15 ms
- Setting up security
- 19 ms
- Waiting for the server
- 19 ms
- Receiving the response: most of the time
- 27 ms
- Total
- 80 ms
Singapore20 ms
- Finding the server
- 3 ms
- Reaching the server
- 3 ms
- Setting up security: most of the time
- 7 ms
- Waiting for the server
- 3 ms
- Receiving the response
- 7 ms
- Total
- 20 ms
Tokyo422 ms
- Finding the server
- 1 ms
- Reaching the server
- 79 ms
- Setting up security
- 81 ms
- Waiting for the server
- 99 ms
- Receiving the response: most of the time
- 163 ms
- Total
- 422 ms
Mumbai321 ms
- Finding the server
- 2 ms
- Reaching the server
- 58 ms
- Setting up security
- 64 ms
- Waiting for the server
- 74 ms
- Receiving the response: most of the time
- 125 ms
- Total
- 321 ms
The phases come from the controlled run of 4 October 2026. Typical and slower times pool the daily requests. 6 of 6 test locations included: Amsterdam, San Francisco, Montreal, Singapore, Tokyo, Mumbai.
Finding the server is measured once per location, so it sits outside these bars. Open a city to see it.
Bars show the typical request, so the parts add up to its total. Why
Wikipedia still feels slow?
- Check Wikimedia’s status page
- Site-wide incidents are posted there. The routing pattern on this page is not an incident; it held on every daily run.
- This is not an article benchmark
- Article pages, the mobile site, search, editing, images and the MediaWiki API are not measured here. The Wikipedia REST API has its own report.
- Which Wikimedia site answers you depends on your network
- Three different Wikimedia addresses across our six cities in the controlled run. Amsterdam and Singapore connected in 3 to 6 ms; San Francisco shared Montreal's address, and Tokyo and Mumbai shared Singapore's, 58 to 79 ms away. A reader in San Francisco on another network may see a very different number.
Response time over the last 29 days
A dot marks a day well above that location’s usual level.
Show the numbers
| Day | Amsterdam | San Francisco | Montreal | Singapore | Tokyo | Mumbai |
|---|---|---|---|---|---|---|
| 11 September 2026 | 16 ms | 406 ms | 85 ms | 18 ms | 421 ms | 343 ms |
| 12 September 2026 | 16 ms | 409 ms | 112 ms | 17 ms | 433 ms | 334 ms |
| 13 September 2026 | 15 ms | 407 ms | 86 ms | 15 ms | 433 ms | 345 ms |
| 14 September 2026 | 15 ms | 412 ms | 86 ms | 17 ms | 408 ms | 343 ms |
| 15 September 2026 | 16 ms | 405 ms | 83 ms | 17 ms | 409 ms | 366 ms |
| 16 September 2026 | 17 ms | 407 ms | 85 ms | 16 ms | 408 ms | 347 ms |
| 17 September 2026 | 18 ms | 418 ms | 87 ms | 1036 ms ▲ | 409 ms | 343 ms |
| 18 September 2026 | 15 ms | 407 ms | 85 ms | 20 ms | 412 ms | 333 ms |
| 19 September 2026 | 16 ms | 417 ms | 85 ms | 15 ms | 465 ms | 332 ms |
| 20 September 2026 | 19 ms | 405 ms | 85 ms | 24 ms | 411 ms | 342 ms |
| 21 September 2026 | 15 ms | 405 ms | 85 ms | 18 ms | 479 ms | 333 ms |
| 22 September 2026 | 16 ms | 405 ms | 86 ms | 16 ms | 483 ms | 346 ms |
| 23 September 2026 | 16 ms | 258 ms | 246 ms ▲ | 16 ms | 464 ms | 332 ms |
| 24 September 2026 | 14 ms | 420 ms | 97 ms | 17 ms | 409 ms | 344 ms |
| 25 September 2026 | 17 ms | 413 ms | 95 ms | 17 ms | 414 ms | 345 ms |
| 26 September 2026 | 16 ms | 412 ms | 85 ms | 18 ms | 410 ms | 331 ms |
| 27 September 2026 | 14 ms | 419 ms | 86 ms | 16 ms | 475 ms | 334 ms |
| 28 September 2026 | 15 ms | 410 ms | 90 ms | 15 ms | 464 ms | 332 ms |
| 29 September 2026 | 14 ms | 379 ms | 84 ms | 16 ms | 438 ms | 326 ms |
| 30 September 2026 | 17 ms | 384 ms | 83 ms | 16 ms | 423 ms | 324 ms |
| 1 October 2026 | 14 ms | 401 ms | 81 ms | 15 ms | 429 ms | 320 ms |
| 2 October 2026 | 14 ms | 388 ms | 86 ms | 17 ms | 422 ms | 324 ms |
| 3 October 2026 | 14 ms | 385 ms | 80 ms | 21 ms | 383 ms | 311 ms |
| 4 October 2026 | 15 ms | 386 ms | 91 ms | 19 ms | 421 ms | 321 ms |
| 5 October 2026 | 13 ms | 385 ms | 81 ms | 17 ms | 420 ms | 324 ms |
| 6 October 2026 | 13 ms | 374 ms | 81 ms | 14 ms | 422 ms | 320 ms |
| 7 October 2026 | 14 ms | 388 ms | 82 ms | 16 ms | 425 ms | 322 ms |
| 9 October 2026 | 17 ms | 404 ms | 87 ms | 14 ms | 418 ms | 344 ms |
| 10 October 2026 | 15 ms | 434 ms | 86 ms | 19 ms | 480 ms | 345 ms |
About this measurement
GET www.wikipedia.org · 65 requests per location over 13 daily runs · 27 September 2026 to 10 October 2026
What we tested · Website document. The www.wikipedia.org language portal, about 120 KB of HTML, served from Wikimedia's own cache layer (ATS).
How LatencyRadar measures response time →Also measured: Wikipedia API →
Technical details
Doesn’t measure: An article page on a language edition, editing, or search.
One Wikipedia surface: the www.wikipedia.org language portal, about 120 KB of HTML that Wikimedia serves from its own cache servers rather than from a commercial CDN. It is the front door a reader types before choosing a language, and it changes rarely, so nearly every request should be a cache hit; the number is close to the cost of reaching Wikimedia's nearest site and pulling 120 KB back. It does not stand for an article page, which is larger, sometimes uncached and served by a language edition, or for search or editing. The Wikipedia REST API has a separate report.
Between 27 September and 10 October 2026 the typical time was 14 ms in Amsterdam and 16 ms in Singapore, 84 ms in Montreal, and 324 ms in Mumbai, 388 ms in San Francisco and 427 ms in Tokyo. The three slow cities were slow on every daily run: San Francisco's median stayed between 374 and 434 ms, Tokyo's between 383 and 480 ms and Mumbai's between 311 and 345 ms. From outside we cannot see how Wikimedia steers each request to one of its sites, but the address each city connected to and how long the connection took together suggest where that site is. Tokyo and Mumbai connected to the address Singapore used, 58 to 79 ms away, which matches the distance to Wikimedia's Singapore site, the nearest it has to either city. San Francisco is the odd one: Wikimedia operates a caching site there, yet our San Francisco worker connected to the address Montreal used, with a 70 ms connection time that suggests the US East Coast. We cannot say from outside whether that routing is Wikimedia's choice, our worker's network provider or something in between; a reader in San Francisco on another network may see a very different number. No city had a bad day; the slowest run for San Francisco, Tokyo and Mumbai was the last one, on 10 October, at 434, 480 and 345 ms. The Google homepage is the other reference site in this set.
| Test location | Typical response time | Slower response time | Requests |
|---|---|---|---|
| Amsterdam | 14 ms | 18 ms | 65 |
| San Francisco | 388 ms | 434 ms | 65 |
| Montreal | 84 ms | 92 ms | 65 |
| Singapore | 16 ms | 24 ms | 65 |
| Tokyo | 427 ms | 516 ms | 65 |
| Mumbai | 324 ms | 345 ms | 65 |
Typical: half of the requests finished within this time (technical: p50). Slower: 95% of requests finished within this time (technical: p95). Statistics
- Request
- GET www.wikipedia.org
- Measured from
- Amsterdam · San Francisco · Montreal · Singapore · Tokyo · Mumbai
- Requests
- Daily measurements from 27 September 2026 to 10 October 2026. 6 of 6 test locations included: Amsterdam, San Francisco, Montreal, Singapore, Tokyo, Mumbai. · 27 September 2026 to 10 October 2026
- Timings taken
- DNS, connect, TLS, waiting for the server, download
- Report coverage
- 6 of 6 test locations included: Amsterdam, San Francisco, Montreal, Singapore, Tokyo, Mumbai.
- Coverage changes
- Mumbai (measured since 7 September 2026) joined this report on 21 September 2026. Earlier daily bands keep their previous location set.
- Daily requests in each location
- Amsterdam: 65 · San Francisco: 65 · Montreal: 65 · Singapore: 65 · Tokyo: 65 · Mumbai: 65
- Typical response time
- Median of the eligible daily requests in each included location.
- Slower response time
- The 95th percentile of the same eligible daily requests in each included location.
Wikipedia
https://www.wikipedia.org/
Independent measurement by LatencyRadar. Not affiliated with Wikipedia. · How we measure (v1.3)
How fast does your site load around the world?
Run a free speed test from multiple cities and find out where your users are waiting. No setup, no account required.
No account required · Takes about 30 seconds.