We explore large-scale training of generative models on video data. Specifically, we train text-conditional diffusion models jointly on videos and images of variable durations, resolutions and aspect ratios. We leverage a transformer architecture that operates on spacetime patches of video and image latent codes. Our largest model, Sora, is capable of generating a minute of high fidelity video.
Recommended
Simplifying, stabilizing, and scaling continuous-time consistency modelsOpenai We’re bringing the Financial Times’ world-class journalism to ChatGPTOpenai Language models are few-shot learnersOpenai Credit Saison achieves 954% ROI in a ChatGPT Enterprise pilotOpenai Making it easier for websites to express AI training preferencesOpenai Community safety report and new protections for teensOpenai
Docs-as-code solutions for API teams: how to choose the right platform in 2026Mintlify
10 best AI observability tools for monitoring and evaluating agents in 2026Mintlify
Video: Building a Live Transcription Badge With Deepgram - Deepgram Blog ⚡️Deepgram
Calling Your Video Game With Your Phone: Part 1Deepgram
Build a Presentation Coaching Application with Recall - Deepgram Blog ⚡️Deepgram
The Most Important Work in AI Training Is Also the Most Overlooked - Deepgram Blog ⚡️Deepgram
New Spanish and Turkish Language Models and Updated General Models - Deepgram Blog ⚡️Deepgram
Workato integrations: What's included and when to use ZapierZapier
AI frameworks: Definition, types, and how to chooseZapier
The Best Video Editing Software in 2026 (Including Free Options)Zapier
Intel Unveils Next Generation Neuromorphic Computing ChipThenewstack
Transform and Future-Proof Your Architecture with MACHThenewstack
Discovered Stacks: One Place for All Your InfrastructurePulumi
How a top tier European soccer team sped up their data processing and reduced costs by 50%Modal
