Inference Space

GPT-5.6 Terra

GPT

Latency & token rates

TTFT (p50)

22,538ms
p50 latency, last 7d

p50 latency, last 7d

Input token rate

5,881token/s

Output token rate

21token/s

Traffic & availability

Requests, 7d

23,545

Requests, 30d

352,164

97.52% uptime over the trailing 7 days.

Availability

97.52%

7d uptime

OperationalDegradedOutage
GPT-5.6 Terra — Performance | Inference Space