ChatGPT API
990msTypical
Fastest of ChatGPT, Claude and Grok.
View benchmarkFind where your site or API is slow, anywhere in the world
Response times from 6 locations worldwide. Free, no account.
No account required.
Fast to run
Paste a URL, run the test, and see how your app responds around the world
Global view
See how your app performs across Europe, America, and Asia
Compare against real services
See how your API performs next to GitHub, Cloudflare, OpenAI, and more
Latest findings
Comparison · 10 Oct
ChatGPT vs Claude vs Grok, who's fastest?
ChatGPT was usually first to respond, Grok usually last.
See comparison →Comparison · 10 Oct
DeepSeek direct or via OpenRouter?
Asia favored direct. Montreal favored OpenRouter.
See comparison →Monthly report · 27 Sept
State of AI API latency, September 2026
Provider choice mattered more than location.
Read the report →
How fast are AI APIs, worldwide?
Daily time to first token from six locations worldwide
Claude API
1132msTypical
Close behind, with less variation by location.
View benchmark
Grok API
2649msTypical
More than twice as slow in most locations.
View benchmarkLive API status
Checked every 30 minutes from six cities
Learn
Practical guides to understanding API latency.
Learn4 min read
What is Time to First Token? (TTFT)
The wait before an AI starts answering, and the point where people notice it.
Learn6 min read
Is 340 ms slow?
It depends on what you're measuring.
Learn5 min read
Test API latency from multiple regions
Measure worldwide latency without running your own infrastructure.