Vincent video model

At the Google I/O Developer Conference 2024, Google introduced Veo, a strong competitor to OpenAI's Sora. Here's a closer look at the product. Veo Highlights Veo is Google DeepMind's text-based video generation model that creates high-quality 1080p videos of more than 60 seconds. It combines text and images to produce videos that meet both input requirements. It also supports video editing using text descriptions, including editing specific areas of the video. About Google Ve...

OpenAI launched the text-to-video Sora three months ago, sparking widespread discussion among netizens, the media, and insiders. At the 2024 I/O Developer Conference held today, Google also launched a benchmark product, Veo, which can generate "high-quality" videos with a length of more than 1 minute and a resolution of up to 1080P, with a variety of visual and film styles. According to Google's official press release, Veo has advanced understanding of natural language and can understand film terms such as "time-lapse photography" and "aerial scenery". Users can use text, images...

Market research firm Factorial Funds recently released a report that OpenAI needs 720,000 Nvidia H100 AI accelerator cards at peak times to deploy its text-to-video model Sora. If each Nvidia H100 AI accelerator card costs $30,000, 720,000 cards would cost $21.6 billion (currently about RMB 156.168 billion). And this is just the cost of the H100 accelerator card, which consumes a lot of electricity to run. Each H100 consumes 700W of power…

Recently, the YouTube channel WVFRM Podcast invited the core team members of OpenAI's video model Sora to participate in an interview. The three members include Bill Peebles, Tim Brooks and Aditya Ramesh, all of whom are Sora project leaders. The three leaders talked about when Sora will be available to users. They said that Sora is still in the feedback acquisition stage, "It is not a product yet, and it will not be released to the public in the short term...

In the early morning of February 16, OpenAI released the innovative text-generated video model - Sora - on its official website. From the effect of Sora-generated video shown by OpenAI on its official website, it is very good in terms of generating video quality, resolution, text-semantic restoration, video action consistency, controllability, details, color, etc.! In particular, it can generate videos of up to 1 minute in length! Surpassing mainstream products such as Gen-2, SVD-XT, Pika, etc., it's a king's bomb right out of the gate. On September 21, 2023, OpenAI released the Vincennes map model DALL-E 3, together with the current Sor...