Study 00 · July 2026

Fixed-stack orchestration benchmark

The original six-platform leaderboard, preserved as benchmark history after the matched 82-scenario Benchmark v1 release became primary.

Historical release. Different cohort and methodology from Benchmark v1; percentages are not directly comparable.

Archived leaderboard

Six platforms · 59 evaluators · 1062 retained calls

pass³ · 3/3 runs passed
RankPlatformpass³pass¹Median latencyReport
01Retell96.6%98.9%1.96sView runs ↗
02Vapi94.9%98.3%2.34sView runs ↗
03Pipecat89.8%95.5%3.15sView runs ↗
04LiveKit84.7%94.9%2.46sView runs ↗
05Synthflow81.4%90.4%3.16sView runs ↗
06ElevenLabs76.3%88.1%1.73sView runs ↗

v0 attempted to hold one agent and its core model and speech components constant. Latency is the archived median per-turn measure; use this page as historical evidence only.