
One API for AI image, video, LLM and audio models, billed only when requests succeed
One API for video, image, and audio models.
Kinovi provides a single API endpoint for running video, image, and audio AI models. It offers pay-as-you-go credits, clear model IDs and rates, and documentation with examples. The same balance can be used in Kinovi Studio.
Kinovi is a platform that lets you run AI models for video, image, and audio generation through one API. It features a catalog of 31+ models from 9 providers, with model-specific inputs and per-second or per-image billing. The API uses a createTask endpoint and polling for results. Kinovi also offers a Studio interface for creation and model discovery, sharing the same balance as the API.
Pay-as-you-go credits with one shared balance. Billing is per second for video and per image for images. Specific rates are listed for models such as Seedance 2.5, Seedance 2.0, and others, with savings compared to Fal.ai.

One API for AI image, video, LLM and audio models, billed only when requests succeed


AI subliminal maker for custom, studio-quality subliminal audio


AI video generator and creative studio for planning, generating and editing video with images, audio and music.

AI speaker separation that splits one recording into individual, time-aligned WAV tracks per detected speaker.

Test your speakers and headphones in seconds โ left, right and stereo channels, a full 20 Hzโ20 kHz sweep, polarity and test tones.


AI podcast studio that turns any content into audio and video podcasts

