Concentrate AI Gemini AI Gateway Benchmarks
Gemini AI Gateway Benchmarks
Last benchmarked August 21, 2026
Request Latency
Cold E2E
2353ms
Median
Warm TTFT
1851ms
Median
Tokens/sec
86
Median
DNS
5.2ms
Median
TCP
1.0ms
Median
TLS
4.5ms
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.
DNS5.2ms
TCP1.0ms
TLS4.5ms
To First Byte1985.6ms
First Byte → Token336.9ms
Performance Over Time
Iteration Distribution
Cold E2E
Warm TTFT
Tokens/sec
DNS
TCP
TLS
Partners