Leaderboard · Model
Alibaba Qwen-TTS
Ranked #7 of 11 models by blind A/B votes against a real human recording. Humanness 100 means listeners can no longer tell it from the person.
Standings as of August 31, 2026 · refreshed every 5 minutes
76
humanness (0–100)
65
live blind rounds
1908ms
warm first-audio
$0.012
per minute (list)
Hear it
“Thanks for calling — I can get that rescheduled for Thursday morning. Does that work for you?”
Background
Alibaba's Qwen3-TTS belongs to the Qwen model family; the arena runs the instruct-flash tier. Strong multilingual coverage, with latency that reflects its data-center distance from US listeners.
On the board
Live results — the numbers move as votes land. Latency is measured at the Speko gateway (warm first-audio, n=30); price is the provider's published rate. Method details on the methodology page; raw standings as open data (CC BY 4.0).