Get a weekly rundown of the latest AI models and research... subscribe!

Deforum Kandinsky 2 2



The model, deforum-kandinsky-2-2, is a text-to-video AI that generates videos basing on text prompts. It takes several inputs including frame rate, seed, steps, video dimensions (width and height), scheduler, animation actions, maximum frames, prompt durations, and animation prompts. The prompts include descriptions such as a scene (e.g., winter forest, spring forest, etc.), specific details (e.g., snowflakes, sun rays, etc.), style, and more. The output of the model is a URL where the generated video can be accessed and downloaded.

Use cases

The deforum-kandinsky-2-2 AI model, designed to generate videos from text prompts, presents a wide range of use cases across various domains. For instance, filmmakers can use it to create initial prototypes of their scenes or animations based on their script. Educators can employ this model to transform textual learning materials into visually stimulating videos, increasing student engagement and learning efficiency. Additionally, marketing teams can conveniently generate promotional videos using descriptive copy content. Specialists in the field of mental health can also use this model to devise therapeutic visuals based on individualized prompts. Potential practical products using this model could include automated video content generation software for social media platforms, personalized video creation apps, or even a therapeutic visualization tool in the field of mental health. Furthermore, video game designers could leverage the model to generate specific game visuals or entire game environments based on text descriptions.



Cost per run
Avg run time

Creator Models

Inst Inpaint$?251
T2i Adapter Sdxl Openpose$?4,645
Lightweight Openpose$?1,442

Similar Models

Try it!

You can use this area to play around with demo applications that incorporate the Deforum Kandinsky 2 2 model. These demos are maintained and hosted externally by third-party creators. If you see an error, message me on Twitter.

Currently, there are no demos available for this model.


Summary of this model and related resources.

Model NameDeforum Kandinsky 2 2
Generate videos from text prompts with Kandinsky-2.2
Model LinkView on Replicate
API SpecView on Replicate
Github LinkView on Github
Paper LinkNo paper link provided


How popular is this model, by number of runs? How popular is the creator, by the sum of all their runs?

Model Rank
Creator Rank


How much does it cost to run this model? How long, on average, does it take to complete a run?

Cost per Run$-
Prediction Hardware-
Average Completion Time-