Models
Our own models and the best of our partners', for video, images, voice, music and understanding. One API key, one MCP server and one credit balance reach all of them, and every one can drop its result straight into an editable project.
Our own image-to-video model. A router sends camera moves and region cinemagraphs to fast depth-parallax and flow pipelines, and faces, bodies, vehicles and anything described in words to a generative tier we run on our own GPU. Every clip is checked so nothing appears that the still does not show.
Vidmoat · Video generation
Bring a still to life: camera moves, moving water and smoke, people breathing and moving. Nothing new is added.
Vidmoat · Editing
Describe an edit in plain words; the agent plans it as editor commands and applies them to the project.
Vidmoat · Transcription
Word-level timings for captions, run on our own servers.
Vidmoat · Media analysis
Follow a person through a clip so text, stickers and crops stay locked to them.
Vidmoat · Media analysis
Cut points, camera and subject motion, and exposure and blur, as numbers and time ranges.
Vidmoat · Media analysis
Find faces and plan a vertical crop that keeps the subject in frame.
Vidmoat · Media analysis
Where people speak, where nothing happens, and whether a track is speech, music or noise.
Vidmoat · Media analysis
The strongest moments of a long video, what it needs before editing, and the text already on screen.
xAI · Video generation
Text or image to video, up to 15 seconds, with optional generated sound.
Alibaba · Video generation
Up to 1080p video from text, a first frame or reference images.
Alibaba · Video generation
Video at 480p, 720p or 1080p, 2 to 15 seconds, with optional audio.
xAI · Image generation
The fast, low-cost route for stills, with up to five reference images.
OpenAI · Image generation
Precise prompt following and legible text in images, at 1k or 2k.
Google · Image generation
Google's image model on Vertex AI, cheaper per reference image.
OpenAI · Image generation
Cut-out stickers with a transparent background, in six styles.
xAI · Speech
Natural voiceover from text, five voices, the lowest price.
OpenAI · Speech
Thirteen voices and delivery you can direct in words.
Google · Music
Original, licence-safe music from a description: a 30-second clip or a full track.
fal.ai · Editing
Remove an object or person from a video and fill the gap convincingly.
Test keys return sample results and spend nothing, so you can build the whole flow before you pay for a single generation.