OpenAI AI Gateway Benchmark
Request Latency
BaselineNo gateway involved — a direct call to Anthropic's API, used as the no-gateway control every other provider is compared against.
Cold E2E
539ms
Median
Warm TTFT
517ms
Median
Tokens/sec
56
Median
DNS
0.5ms
Median
TCP
1.1ms
Median
TLS
3.9ms
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.
DNS0.5ms
TCP1.1ms
TLS3.9ms
To First Byte302.0ms
First Byte → Token225.5ms
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.