Anthropic AI Gateway Benchmark
Request Latency
Cold E2E
659ms
Median
Warm TTFT
505ms
Median
Tokens/sec
56
Median
DNS
2.3ms
Median
TCP
1.0ms
Median
TLS
7.6ms
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.
DNS2.3ms
TCP1.0ms
TLS7.6ms
To First Byte635.4ms
First Byte → Token0.1ms
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.