Unverified50% confidenceReleaseExact time
Teams from Fudan University, Tencent Hunyuan, and Zhejiang University open-sourced the audio-video joint generation framework Prism, which can natively generate videos with sound at resolutions from 720P and 1080P to 2K.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/8/2026
First Seen
Valid until: 1/6/2027
Sources
Related Entities
Related Claims
UnverifiedPrism allows users to freely create audio visualization effects.68% similarUnverifiedThe Blender reference video was rendered at a resolution of 1080×1080, 30fps, with a duration of five seconds.67% similarUnverifiedCurrent mainstream video processing uses a static approach in which the model ingests video content frame by frame at a fixed frame rate (defaulting to 1 FPS, adjustable via API).67% similarUnverifiedKling AI announced that its video generation model 4.0 will officially launch in October; 4.0 Flash has already been made available to a limited number of Black Gold annual card members, with the company stating the new version supports native 30-second video output, up to 15 multimodal references, and 10 keyframe inputs.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/989147API
curl https://kongchang.com/api/v1/knowledge/claims/989147MCP
get_claim(id=989147)