OpenAI AI Gateway Benchmark
Request Latency
Cold E2E
976ms
Median
Warm TTFT
668ms
Median
Tokens/sec
47
Median
DNS
1.6ms
Median
TCP
14.4ms
Median
TLS
17.3ms
Median
Cold Connection Breakdown
Where a fresh request's time goes: DNS lookup, TCP connect, TLS handshake, then time to first response byte, then first byte to first streamed token.
DNS1.6ms
TCP14.4ms
TLS17.3ms
To First Byte530.6ms
First Byte → Token333.2ms
Performance Over Time
Iteration Distribution
Cold E2E
Warm TTFT
Tokens/sec
DNS
TCP
TLS
Become a partner
Want your logo featured across our benchmarks? Become a ComputeSDK benchmark partner.