OpenAI Responses: GPT-5.3 Codex
Service quality measured from 22 regions around the world
Standard: The chat completions API sends the full prompt and waits for the complete reply in one response, the usual non-streaming chat completions call.
Average response time
1.2s
Compare regions
Line chart showing response time over 30 days for 2 regions. 402 data points loaded.
Regional measurements
Africa
| Region | Response time | Status | Alerts |
|---|---|---|---|
| Cape Town | 1.5s |
Asia Pacific
| Region | Response time | Status | Alerts |
|---|---|---|---|
| Hong Kong | - | ||
| Jakarta | 2.0s | ||
| Mumbai | 1.3s | ||
| Osaka | 1.4s | ||
| Seoul | 1.1s | ||
| Singapore | 1.0s | ||
| Sydney | 1.3s | ||
| Tokyo | 948ms |
Europe
| Region | Response time | Status | Alerts |
|---|---|---|---|
| Dublin | 1.2s | ||
| Frankfurt | 755ms | ||
| London | 4.6s | ||
| Milan | 953ms | ||
| Paris | 866ms | ||
| Stockholm | 1.2s |
Middle East
| Region | Response time | Status | Alerts |
|---|---|---|---|
| Bahrain | - |
North America
| Region | Response time | Status | Alerts |
|---|---|---|---|
| Montreal | 913ms | ||
| N. California | 1.2s | ||
| N. Virginia | 1.0s | ||
| Ohio | 728ms | ||
| Oregon | 1.6s |
South America
| Region | Response time | Status | Alerts |
|---|---|---|---|
| São Paulo | 1.5s |