As video becomes central to how brands communicate, train, and educate audiences, CraftStory has introduced a new capability aimed at removing long standing production barriers in human led content. CraftStory today announced the release of its Image to Video model, a major expansion of its Model 2.0 platform that allows users to generate up to five minute, studio quality human videos using only a single image and a written script.
The launch marks a significant step forward for AI video creation in the MarTech and enterprise content ecosystem. While many AI tools can generate short clips, sustaining realism, motion, and identity over long durations has remained a challenge. CraftStory’s Image to Video technology addresses that gap by enabling expressive, multi minute performances without requiring source footage or complex production workflows.
CraftStory first entered the market in November 2025 with a Video to Video model that animated still images using motion captured from a driving video. With Model 2.0, the company removes that dependency entirely. Businesses can now create long form human videos starting from a photo and text alone, while maintaining the same realism, continuity, and performance quality previously limited to video driven workflows.
Marketing Technology Insights: Teads Expands Executive Leadership Team With Appointment of Dani Cushion as CMO
The Image to Video model is designed to support use cases such as product explainers, employee training, customer communication, and educational content. Users upload a single image of a person along with a script or audio track. CraftStory’s AI then synthesizes a complete video performance, generating realistic lip sync, facial expressions, gestures, and body language that evolve naturally over time and remain consistent throughout the video.
At a technical level, Model 2.0 shares the same core architecture as CraftStory’s video based system. It includes high fidelity lip sync, identity preservation across long sequences, and advanced gesture generation algorithms that infer appropriate movement directly from speech rhythm and emotional tone. The system also animates scene motion, creating a more immersive and believable result than static talking head videos.
In addition, CraftStory is introducing support for moving cameras. The platform can now generate walk and talk videos up to 80 seconds long, where the subject moves through a scene while the camera tracks naturally. This feature, currently in beta, allows teams to produce more dynamic, cinematic shots that stand apart from traditional on camera formats.
Marketing Technology Insights: Affle Appoints Sameer Sondhi as Chief Strategic Investments Officer and CEO, North America
Underlying Model 2.0 is a proprietary parallelized diffusion pipeline built to scale long form video generation. By processing multiple temporal segments simultaneously while enforcing global coherence, CraftStory maintains visual consistency across minutes of footage. The model was trained on high frame rate recordings of real actors, capturing subtle facial dynamics and expressive body motion to avoid robotic outputs.
“Image to Video is a major step toward fully script driven video creation,” said Victor Erukhimov, Founder and CEO of CraftStory. “You no longer need to record a video to get a realistic human performance. If you have an image and something to say, Model 2.0 can turn that into a believable, long form video.”
Looking ahead, CraftStory plans to advance Model 2.0 toward fully automated text to video workflows. The company’s focus is on making high quality human video creation faster, more scalable, and more accessible for marketing, training, and enterprise communication teams.
Marketing Technology Insights: Marketing Architects Partners With Jounce Media for Greater Transparency in CTV Buying
For media inquiries, you can write to our MarTech Newsroom at info@intentamplify.com
