HPCAI Tech has released Open-Sora 1.2, an open-source video generation model now available on Hugging Face. The model, which has 1.1 billion parameters, was trained on over 30 million data points and can generate up to 16-second, 720p HD videos. Open-Sora 1.2 includes features such as a video compression network and rectified-flow training, aiming to make high-quality video production more accessible. The model's capabilities are showcased in a GitHub repository, allowing users to generate short films with ease.
🌟 Witness the magic of our text-to-video model! 🎥 Open-Sora @HPCAITech is revolutionizing content creation. Watch as we transform text into stunning videos, empowering creators to bring their ideas to life. Explore our GitHub repository for a glimpse into the future of… https://t.co/xhmAm9rw8U
Best open-source video generation model? It probably is and it’s not even close. I’ve been closely tracking the Open-Sora project, from HPC-AI Tech, and their new release is impressive. You can now generate 16-second 720p videos with Open-Sora! Includes: > 1.1B model on >30… https://t.co/TiHZlGX0Ek
Open-Sora: You can generate any 16-second 720p HD video with one click. This is awesome! The model is completely open source. You can use the model weights to generate short films. Here is the GitHub repo: https://t.co/TGuk6sBZ4Q Few examples below: https://t.co/JychbsNTwM
Open-Sora 1.2 is out🔥 Open-Sora is an initiative dedicated to efficiently producing high-quality video in open-source way , released by @HPCAITech 👏 Model: https://t.co/9gJB6iLGLC Demo: https://t.co/npE3DeXbFo ✨ Video compression network ✨Rectifie-flow training ✨More data…
What a week for video generation! After the hype around closed-source models from Kling, Luma, and Runway, Here comes the open-source one: Open-Sora 1.2 from @HPCAITech just got released @huggingface Highlights: 1.1B model, trained on >20M data, generate up to 14s, 720p video https://t.co/tn3bVM5wj1