All tasks / Models

PrismAudio

Alibaba's open video-to-audio line; PrismAudio, released in the ThinkSound repo, uses chain-of-thought reasoning and RL for video sound.

Alibaba (FunAudioLLM)
Open weights
Released 24 Mar 2026

Where it ranks

–
Sound effects
Models · Local · no result on this task’s benchmarks yet

Facts

Vendor: Alibaba (FunAudioLLM)
Family: ThinkSound
Released: 24 Mar 2026
Weights: Open: download and run it yourself
Access: Free (open weights)
Last verified: 18 Oct 2026

Price from github.com, 18 Oct 2026.

Benchmark results

No public benchmark results recorded yet.

Where you can use it

No tracked tool lists this model yet; as open weights it runs in local runtimes.

Earlier versions

ThinkSound · Jul 2025

Sources