Skip to content
GitHub
Compute SDK

Product

Benchmarks

Resources

Docs Blog Become a partner

Anthropic (Direct) AI Gateway Benchmarks

Anthropic (Direct)
AI Gateway Benchmarks

Last benchmarked July 24, 2026

No-gateway control — a direct call to Anthropic's API, used to measure how much latency each gateway adds.

Request Latency

Baseline
100% success

No gateway involved — a direct call to Anthropic's API, used as the no-gateway control every other provider is compared against.

Cold E2E
821ms
Median
Warm TTFT
521ms
Median
Tokens/sec
88
Median
DNS
0.8ms
Median
TCP
1.1ms
Median
TLS
2.9ms
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.

DNS0.8ms
TCP1.1ms
TLS2.9ms
To First Byte810.3ms
First Byte → Token0.3ms

Performance Over Time

Iteration Distribution

Cold E2E

Warm TTFT

Tokens/sec

DNS

TCP

TLS

PartnersLatitude