imageat

imageat is a generative video tool. A multi-model video studio that puts Veo, Kling, Seedance, Wan, FLUX, MiniMax, PixVerse and Grok behind one prompt bar, with text, image, first and last frame, motion transfer and extend modes, plus API and MCP access.

imageat's AI Video Generator is an aggregator studio rather than a single model. One prompt bar sits in front of Google Veo 3.1, Kling 3 and Kling 2.6, ByteDance Seedance 2.0 and OmniHuman v2, Black Forest Labs FLUX 3 Max, Alibaba Wan 3.0 and Happy Horse 1.0, MiniMax H3 Max, PixVerse V6, Grok Imagine Video, plus Genjutsu Light for motion work. Instead of subscribing to each vendor separately, users pick a model from a dropdown, set aspect ratio, duration, resolution and audio, and spend credits only when a render actually runs. Kling 4.0 is announced as coming soon once its API opens. The workflow covers more than prompt-to-clip. Text to video starts from a written description; image to video animates a still while holding the subject consistent; first and last frame mode interpolates a transition between two uploaded images. OmniHuman v2 pairs one photo with an MP3 or WAV file to produce a talking or singing presenter. Genjutsu Light transfers motion from a source clip onto reference characters or swaps an object using reference images. PixVerse V6 extends existing footage, and Grok's edit mode rewrites a clip from a text instruction. Output reaches 1080p with durations that vary by model, from a few seconds up to 30 on Wan 3.0 and Kling 4.0. Practically, the tool suits social media producers, marketers, indie filmmakers and agencies that need to test several engines without stitching vendors together. A typical run takes two to five minutes, which is fine for iteration but slow for live work. Teams with engineering capacity can skip the interface: an API and an MCP server let products and AI agents trigger renders and receive CDN URLs, and there is a prompt library plus trending style pages for people who do not want to write from scratch. Against single-model tools, the draw is breadth and per-use billing; the tradeoff is that quality and limits shift model by model, so the studio rewards users who read the specs before spending credits.