class B live

Alibaba's joint video+audio model; launched anonymously, topped the AA arena Apr 2026, now #2 text-to-video behind Seedance 2.0. Closed, API via aggregators.

Layer 1 — factual axes

API & price public via aggregators (fal.ai ~$0.14-0.28/s); no first-party endpoint secondary source ↗ 2026-07-04
Max output 3-15s · 1080p (24fps) secondary source ↗ 2026-07-04
Native audio yes (joint video+audio, multilingual lip-sync) secondary source ↗ 2026-07-04
Weights closed (vendor states no open-source) secondary source ↗ 2026-07-04
Free tier no (paid API only) secondary source ↗ 2026-07-04

Layer 2 — output quality

1152 / 1600
Eligible for Best for APINative audio

Alibaba’s joint video+audio model; launched anonymously, topped the AA arena Apr 2026, now #2 text-to-video behind Seedance 2.0. Closed, API via aggregators.

Data as of 2026-07-04 · verified 2026-07-04. One ruler for all tools. How we measure ↗