CraftStory, a new AI startup founded by the creators of OpenCV, has launched a video generation system called Model 2.0, capable of producing realistic, human-centric videos up to five minutes long. This marks a significant advancement in the AI video space, where competitors like OpenAI's Sora and Google's Veo typically max out at 25 seconds or less. The system aims to address the limitations of current models, which often fail to deliver longer, coherent video sequences necessary for corporate training and marketing needs. CraftStory's technology relies on a parallelized diffusion architecture that allows multiple smaller algorithms to run simultaneously across the video's entire duration. This contrasts with traditional methods that require larger networks and more resources for longer clips. The company trained its model on proprietary footage, ensuring high-quality video output without the need for vast amounts of internet-sourced data. Users can upload still images and driving videos to create clips, with advanced features for lip-syncing and gesture alignment enhancing realism. With $2 million in funding, primarily from Andrew Filev, CraftStory faces a challenging landscape dominated by competitors with billions in capital. However, Erukhimov, the CEO, argues that success doesn't solely depend on funding but on the quality of the technology and the team's expertise. CraftStory is targeting enterprise applications specifically, positioning itself to fill gaps in long-form video content for training and product demonstrations. The company believes that their approach can save businesses significant time and money compared to traditional video production methods.