Benchmark report
Is the Jev API slow?
Jev response times from 6 cities, measured daily
Mixed by location
The full answer typically came back in 99 ms from San Francisco and 287 ms from Mumbai, and no day in the window was slow in every city.
- Fastest
- 99 ms
- San Francisco
- Slowest
- 287 ms
- Mumbai
- Successful requests
- 100%
- 6 locations
Response time around the world
San Francisco
99 msMontreal
117 msTokyo
160 msAmsterdam
229 msSingapore
243 msMumbai
287 ms
Why does Mumbai wait three times longer than San Francisco?
In the controlled run of 4 October, connecting and the secure handshake took under 25 ms in every city, each at a nearby Cloudflare site. Almost all of the time was waiting for the answer: 75 ms from San Francisco, 125 ms from Montreal, 156 ms from Tokyo, 214 ms from Amsterdam, 242 ms from Singapore and 292 ms from Mumbai. The wait grows with distance from the US West Coast, which suggests Jev answers from there and the other cities add the trip to it.
Click the location to see each stage.
Mumbai304 ms
- Finding the server
- 2 ms
- Reaching the server
- 3 ms
- Setting up security
- 9 ms
- Waiting for the server: most of the time
- 292 ms
- Receiving the response
- 0 ms
- Total
- 304 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.
Amsterdam235 ms
- Finding the server
- 171 ms
- Reaching the server
- 8 ms
- Setting up security
- 13 ms
- Waiting for the server: most of the time
- 214 ms
- Receiving the response
- 0 ms
- Total
- 235 ms
San Francisco91 ms
- Finding the server
- 14 ms
- Reaching the server
- 4 ms
- Setting up security
- 12 ms
- Waiting for the server: most of the time
- 75 ms
- Receiving the response
- 0 ms
- Total
- 91 ms
Montreal133 ms
- Finding the server
- 2 ms
- Reaching the server
- 3 ms
- Setting up security
- 4 ms
- Waiting for the server: most of the time
- 125 ms
- Receiving the response
- 1 ms
- Total
- 133 ms
Singapore260 ms
- Finding the server
- 6 ms
- Reaching the server
- 6 ms
- Setting up security
- 11 ms
- Waiting for the server: most of the time
- 242 ms
- Receiving the response
- 1 ms
- Total
- 260 ms
Tokyo166 ms
- Finding the server
- 1 ms
- Reaching the server
- 3 ms
- Setting up security
- 6 ms
- Waiting for the server: most of the time
- 156 ms
- Receiving the response
- 1 ms
- Total
- 166 ms
Mumbai304 ms
- Finding the server
- 2 ms
- Reaching the server
- 3 ms
- Setting up security
- 9 ms
- Waiting for the server: most of the time
- 292 ms
- Receiving the response
- 0 ms
- Total
- 304 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
TypeSafe still feels slow?
- Check TypeSafe’s status page
- API incidents are posted there. A slow answer during one is not your network.
- This is a full answer, not a first token
- Jev does not stream, so we time the whole JSON response. The chat model pages time the first visible word of a streamed reply, which is a different measurement.
- The slow first week
- On 21, 22, 24 and 25 September every city was slower than it has been since, most of them about twice as slow: Amsterdam's daily median was 475 to 554 ms on those days and 225 to 289 ms from 26 September on. Those days come before this page's window, so the figures here leave them out; the history chart still shows them.
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 |
|---|---|---|---|---|---|---|
| 21 September 2026 | 537 ms ▲ | 194 ms | 315 ms ▲ | 738 ms ▲ | 475 ms ▲ | 812 ms ▲ |
| 22 September 2026 | 554 ms ▲ | 175 ms | 365 ms ▲ | 697 ms ▲ | 503 ms ▲ | 804 ms ▲ |
| 23 September 2026 | 254 ms | 169 ms | 229 ms | 286 ms | 188 ms | 353 ms |
| 24 September 2026 | 553 ms ▲ | 229 ms ▲ | 373 ms ▲ | 718 ms ▲ | 512 ms ▲ | 785 ms ▲ |
| 25 September 2026 | 475 ms | 135 ms | 303 ms ▲ | 656 ms ▲ | 459 ms ▲ | 765 ms ▲ |
| 26 September 2026 | 243 ms | 105 ms | 142 ms | 287 ms | 179 ms | 334 ms |
| 27 September 2026 | 257 ms | 94 ms | 132 ms | 286 ms | 189 ms | 316 ms |
| 28 September 2026 | 251 ms | 123 ms | 195 ms | 273 ms | 193 ms | 335 ms |
| 29 September 2026 | 245 ms | 113 ms | 166 ms | 245 ms | 201 ms | 306 ms |
| 30 September 2026 | 289 ms | 94 ms | 140 ms | 248 ms | 173 ms | 298 ms |
| 1 October 2026 | 234 ms | 87 ms | 138 ms | 281 ms | 179 ms | 303 ms |
| 2 October 2026 | 255 ms | 89 ms | 132 ms | 243 ms | 170 ms | 280 ms |
| 3 October 2026 | 259 ms | 93 ms | 128 ms | 282 ms | 165 ms | 315 ms |
| 4 October 2026 | 239 ms | 84 ms | 146 ms | 257 ms | 173 ms | 325 ms |
| 5 October 2026 | 230 ms | 111 ms | 130 ms | 258 ms | 160 ms | 302 ms |
| 6 October 2026 | 243 ms | 102 ms | 154 ms | 256 ms | 175 ms | 302 ms |
| 7 October 2026 | 236 ms | 119 ms | 134 ms | 251 ms | 183 ms | 306 ms |
| 9 October 2026 | 225 ms | 109 ms | 128 ms | 278 ms | 164 ms | 308 ms |
| 10 October 2026 | 251 ms | 88 ms | 130 ms | 237 ms | 157 ms | 278 ms |
| 11 October 2026 | 235 ms | 98 ms | 146 ms | 233 ms | 181 ms | 286 ms |
| 12 October 2026 | 269 ms | 89 ms | 123 ms | 264 ms | 196 ms | 283 ms |
| 13 October 2026 | 239 ms | 140 ms | 138 ms | 260 ms | 163 ms | 320 ms |
| 14 October 2026 | 208 ms | 118 ms | 104 ms | 243 ms | 140 ms | 278 ms |
| 15 October 2026 | 216 ms | 113 ms | 94 ms | 230 ms | 147 ms | 276 ms |
| 16 October 2026 | 210 ms | 73 ms | 90 ms | 247 ms | 147 ms | 279 ms |
| 17 October 2026 | 225 ms | 86 ms | 99 ms | 235 ms | 155 ms | 269 ms |
| 18 October 2026 | 218 ms | 75 ms | 95 ms | 229 ms | 147 ms | 274 ms |
| 19 October 2026 | 261 ms | 73 ms | 68 ms | 231 ms | 141 ms | 291 ms |
| 20 October 2026 | 214 ms | 101 ms | 69 ms | 240 ms | 152 ms | 301 ms |
About this measurement
POST api.typesafe.ai/v1/systemone · 65 requests per location over 13 daily runs · 7 October 2026 to 20 October 2026
What we tested · AI decision. Whole JSON response from jev-1.13.0 to one yes/no question on a two-sentence support message: POST api.typesafe.ai/v1/systemone, no stream, paid key.
How LatencyRadar measures response time →
Technical details
Doesn’t measure: Time to first token (Jev has none), the Requesty or OpenRouter routes, Cloudflare Workers AI, longer states, more questions or the SDK retry layer.
One TypeSafe surface: POST api.typesafe.ai/v1/systemone with the jev-1.13.0 model, a two-sentence support message and one yes/no question, paid key. Jev returns its answers as one JSON document instead of a stream of tokens, so the number here is the time to the complete answer on a new connection. Your code waits that long before it can act on the decision.
Between 7 and 20 October 2026 the typical time was 99 ms in San Francisco, 117 ms in Montreal, 160 ms in Tokyo, 229 ms in Amsterdam, 243 ms in Singapore and 287 ms in Mumbai. Daily medians stayed between 73 and 140 ms in San Francisco and between 269 and 320 ms in Mumbai, and no day was slow in every city. This page does not rank Jev against the chat models, which measure something different.
| Test location | Typical response time | Slower response time | Requests |
|---|---|---|---|
| Amsterdam | 229 ms | 274 ms | 65 |
| San Francisco | 99 ms | 140 ms | 65 |
| Montreal | 117 ms | 167 ms | 65 |
| Singapore | 243 ms | 301 ms | 65 |
| Tokyo | 160 ms | 201 ms | 65 |
| Mumbai | 287 ms | 342 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
- POST api.typesafe.ai/v1/systemone
- Measured from
- Amsterdam · San Francisco · Montreal · Singapore · Tokyo · Mumbai
- Requests
- Daily measurements from 7 October 2026 to 20 October 2026. 6 of 6 test locations included: Amsterdam, San Francisco, Montreal, Singapore, Tokyo, Mumbai. · 7 October 2026 to 20 October 2026
- Model
- jev-1.13.0, pinned (not jev-latest) · 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: 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.
Jev API
https://api.typesafe.ai/v1/systemone
Independent measurement by LatencyRadar. Not affiliated with TypeSafe. · How we measure (v1.3)
How fast is your API compared to Jev?
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.