Anthropic AI Gateway Benchmark
Request Latency
Cold E2E
695ms
Median
Warm TTFT
617ms
Median
Tokens/sec
49
Median
DNS
1.2ms
Median
TCP
13.9ms
Median
TLS
17.1ms
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.
DNS1.2ms
TCP13.9ms
TLS17.1ms
To First Byte621.5ms
First Byte → Token1.4ms
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.