Benchmark report
Is Cloudflare Clef slow?
Response times from 6 cities
Slow from some locations
A full yes/no answer typically came back in 248 ms from Montreal and 647 ms from Mumbai.
- Fastest
- 248 ms
- Montreal
- Slowest
- 647 ms
- Mumbai
- Successful requests
- 100%
- 6 locations
Response time around the world
Montreal248 ms
Amsterdam278 ms
San Francisco283 ms
Tokyo405 ms
Singapore453 ms
Mumbai647 ms
Why does Mumbai wait more than twice as long as Montreal?
Every city reached a nearby Cloudflare address, and connecting plus the secure handshake took under 20 ms everywhere. Almost all of the time was waiting for the answer, so the gap between cities comes from running the model, not from reaching Cloudflare. Mumbai was also the least steady city: its daily median ranged from 405 to 979 ms.
Click the location to see each stage.
Mumbai434 ms
Mumbai- Finding the server
- 2 ms
- Reaching the server
- 3 ms
- Setting up security
- 7 ms
- Waiting for the server: most of the time
- 422 ms
- Receiving the response
- 2 ms
- Total
- 434 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.
Amsterdam222 ms
Amsterdam- Finding the server
- 1 ms
- Reaching the server
- 8 ms
- Setting up security
- 11 ms
- Waiting for the server: most of the time
- 203 ms
- Receiving the response
- 0 ms
- Total
- 222 ms
San Francisco349 ms
San Francisco- Finding the server
- 12 ms
- Reaching the server
- 5 ms
- Setting up security
- 9 ms
- Waiting for the server: most of the time
- 335 ms
- Receiving the response
- 0 ms
- Total
- 349 ms
Montreal214 ms
Montreal- Finding the server
- 20 ms
- Reaching the server
- 2 ms
- Setting up security
- 5 ms
- Waiting for the server: most of the time
- 207 ms
- Receiving the response
- 0 ms
- Total
- 214 ms
Singapore382 ms
Singapore- Finding the server
- 3 ms
- Reaching the server
- 3 ms
- Setting up security
- 6 ms
- Waiting for the server: most of the time
- 372 ms
- Receiving the response
- 1 ms
- Total
- 382 ms
Tokyo611 ms
Tokyo- Finding the server
- 2 ms
- Reaching the server
- 2 ms
- Setting up security
- 7 ms
- Waiting for the server: most of the time
- 601 ms
- Receiving the response
- 1 ms
- Total
- 611 ms
Mumbai434 ms
Mumbai- Finding the server
- 2 ms
- Reaching the server
- 3 ms
- Setting up security
- 7 ms
- Waiting for the server: most of the time
- 422 ms
- Receiving the response
- 2 ms
- Total
- 434 ms
The phases come from the daily run of 20 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
Cloudflare still feels slow?
- Check Cloudflare's status page
- Workers AI incidents are posted there. A slow answer during one is not your network.
- This is a full answer, not a first token
- Clef does not stream, so we time the whole JSON response, which is a different measurement from the first-token times on the chat model pages.
- We measure clef, not clef-flash
- Clef comes in two sizes. These numbers are for the 27B clef model; the smaller clef-flash is not measured here.
Response time over the last 9 days
Show the numbers
| Day | Amsterdam | San Francisco | Montreal | Singapore | Tokyo | Mumbai |
|---|---|---|---|---|---|---|
| 11 October 2026 | 291 ms | 291 ms | 225 ms | 773 ms | 377 ms | 979 ms |
| 12 October 2026 | 333 ms | 316 ms | 261 ms | 646 ms | 404 ms | 802 ms |
| 13 October 2026 | 276 ms | 209 ms | 261 ms | 326 ms | 291 ms | 785 ms |
| 14 October 2026 | 222 ms | 332 ms | 225 ms | 438 ms | 499 ms | 405 ms |
| 15 October 2026 | 355 ms | 279 ms | 248 ms | 316 ms | 369 ms | 527 ms |
| 16 October 2026 | 519 ms | 277 ms | 419 ms | 471 ms | 336 ms | 603 ms |
| 17 October 2026 | 514 ms | 295 ms | 456 ms | 488 ms | 533 ms | 417 ms |
| 18 October 2026 | 221 ms | 283 ms | 163 ms | 607 ms | 603 ms | 647 ms |
| 19 October 2026 | 179 ms | 245 ms | 182 ms | 441 ms | 514 ms | 590 ms |
| 20 October 2026 | 222 ms | 349 ms | 214 ms | 382 ms | 611 ms | 434 ms |
About this measurement
POST api.cloudflare.com/client/v4/accounts/f14055de773af1c8e9f127a2456dc5e6/ai/run/@cf/cloudflare/clef · 50 requests per location over 10 daily runs · 11 October 2026 to 20 October 2026
What we tested · AI decision. Whole JSON response from Cloudflare's clef to one yes/no question on a two-sentence support message: POST Workers AI /ai/run/@cf/cloudflare/clef, no streaming.
How LatencyRadar measures response time →Also measured: Cloudflare website →
Technical details
Doesn’t measure: Time to first token, other Workers AI models, choice or score questions, or longer inputs.
One Cloudflare surface: Clef on Workers AI, POST api.cloudflare.com/client/v4/accounts/<id>/ai/run/@cf/cloudflare/clef with the clef model, a two-sentence support message and one yes/no question. It runs on our account's free Workers AI allowance. Clef returns its answer as one JSON document, so the number here is the time to the complete answer on a new connection. The Cloudflare page measures how fast cloudflare.com loads.
Between 11 and 20 October 2026 the typical time was 248 ms in Montreal, 278 ms in Amsterdam, 283 ms in San Francisco, 405 ms in Tokyo, 453 ms in Singapore and 647 ms in Mumbai. 95% of requests finished within 516 ms in San Francisco and within 1,509 ms in Singapore. San Francisco was the steadiest city, with daily medians between 209 and 349 ms.
| Test location | Typical response time | Slower response time | Requests |
|---|---|---|---|
| Amsterdam | 278 ms | 825 ms | 50 |
| San Francisco | 283 ms | 516 ms | 50 |
| Montreal | 248 ms | 564 ms | 50 |
| Singapore | 453 ms | 1509 ms | 50 |
| Tokyo | 405 ms | 1176 ms | 50 |
| Mumbai | 647 ms | 1169 ms | 50 |
Typical: half of the requests finished within this time (technical: p50). Slower: 95% of requests finished within this time (technical: p95). Statistics
- Request
- POST api.cloudflare.com/client/v4/accounts/f14055de773af1c8e9f127a2456dc5e6/ai/run/@cf/cloudflare/clef
- Measured from
- Amsterdam · San Francisco · Montreal · Singapore · Tokyo · Mumbai
- Requests
- Daily measurements from 11 October 2026 to 20 October 2026. 6 of 6 test locations included: Amsterdam, San Francisco, Montreal, Singapore, Tokyo, Mumbai. · 11 October 2026 to 20 October 2026
- Model
- clef (27B), on Workers AI · no streaming · whole JSON response timed
- Request
- One yes/no question ("The customer is explicitly asking for a refund") about a two-sentence support message asking for a duplicate charge to be refunded. Identical in every city and on every day
- 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.
- Daily requests in each location
- Amsterdam: 50 · San Francisco: 50 · Montreal: 50 · Singapore: 50 · Tokyo: 50 · Mumbai: 50
- 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.
Cloudflare Clef
https://api.cloudflare.com/client/v4/accounts/f14055de773af1c8e9f127a2456dc5e6/ai/run/@cf/cloudflare/clef
Independent measurement by LatencyRadar. Not affiliated with Cloudflare. · How we measure (v1.3)
How fast is your API compared to Cloudflare Clef?
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.
