Netflix has released VOID (Video Object and Interaction Deletion), its first public AI model. The open-source tool removes objects from video while preserving physically plausible interactions.
Tencent has released OmniWeaving, a unified video generation model that handles seven distinct tasks from text-to-video to reasoning-augmented generation, with publicly available weights and code.
Skywork AI releases Matrix-Game-3.0, an open-source model that generates real-time interactive video at 720p and 40fps.
Meta releases SAM 3.1 with object multiplexing that tracks 16 objects simultaneously at 32fps on a single H100 GPU.
Snapchat launched AI Clips on March 24, a new Lens format that transforms a single photo into a 5-second AI-generated video for Lens+ subscribers.
DeepBrain AI launched Interactive AI Video Agents on March 24, bringing real-time conversational avatars to enterprise workflows with sub-1.5s response time.
Researchers released daVinci-MagiHuman, a 15-billion-parameter open-source model that generates realistic human video with synchronized speech in two seconds.
Seven of the top ten text-to-video models on HuggingFace are Wan2.2 variants, with combined monthly downloads exceeding 500,000. Open-source video generation is no longer catching up to proprietary tools.
Corridor Digital co-founder Niko Pueringer releases CorridorKey, an AI chroma keyer that handles hair, motion blur, and transparency at VFX-production quality.
Adobe and NVIDIA announce a strategic partnership to build next-gen Firefly models on CUDA-X and Cosmos, plus a 3D digital twin solution now in public beta.