InVideo Agent One generates 1-minute anime shorts from one prompt
Creators showed Agent One generating images, video, audio, and a full edit for a minute-long anime sequence from a single prompt, then refining it conversationally with scripts and references. A separate short-film demo said the tool also handled character creation, storyboards, and guided editing, so test it for end-to-end concept work.

TL;DR
- CharaspowerAI's anime demo said InVideo Agent One generated a minute-plus anime sequence from a single prompt, covering images, video, audio, and the final edit.
- In a follow-up workflow post, CharaspowerAI said creators can upload a script, visual references, and creative direction, then keep refining the project through conversation while the agent handles generation and editing.
- Anima_Labs' cat-chaos short and Anima_Labs' Anti-Hero film both framed Agent One as more than a clip generator, crediting it with character creation, script rework, storyboards, animation, and editing.
- InVideo's Create with Agent One help collection describes the product as a chat-based system that "builds, edits and thinks like a filmmaker," while the pricing page says paid plans include access to an InVideo v4 agent that can create up to 30 minutes of video from one prompt.
You can watch the anime sequence, jump to a comedic short film, and compare those creator posts with InVideo's own help center hub, which lists docs for context, prompt guides, credits, and a masterclass for brand films. A separate official workflow post shows the same system pitched for product swaps, localization, and brand-rule memory, while Kingy AI's review zeroes in on the big trick: persistent project context.
One-prompt anime sequence
The cleanest claim in the evidence pool is simple: one prompt in, one minute of anime out. In CharaspowerAI's post, the creator said Agent One generated the images, videos, audio, and full cut, with the human role narrowed to direction and a few refinements.
That matches InVideo's broader homepage pitch at InVideo, where the product is framed as a system for turning an idea into a finished video and then editing it conversationally. The interesting part is not raw generation alone, it is the collapse of storyboard, asset creation, soundtrack, and assembly into one chat loop.
Context and conversation
According to CharaspowerAI, the front door is a bundle of context, not a tiny prompt. The creator listed three inputs:
- a script
- visual references
- creative directions
From there, the workflow post says Agent One handles images, video, audio, and editing, then stays open as a conversational editor for scene changes and new directions. InVideo's Create with Agent One help collection uses nearly the same framing, calling it a system that "builds, edits and thinks like a filmmaker," and linking docs for context, prompt guides, credits, and the difference from Autopilot.
In an official blog post about scaling ads, InVideo extends that idea from filmmaking to production ops. The company says Agent One can keep brand context, swap 100-plus products, and localize the same base ad across languages and regions. A separate official Runway guide says creators can "skip prompt engineering" and let Agent One structure shot logic for them.
Short-film demos
The two strongest companion demos come from Anima_Labs, and both widen the product story beyond anime clips. In the cat-meeting short, the creator said Agent One helped with character creation, storytelling, storyboards, video production, and editing. In the Anti-Hero post, Anima_Labs said the tool reworked ideas, script, and characters, then generated a shot-by-shot storyboard and the animation before handling the edit.
That is a more ambitious claim than "text to video." It positions Agent One as a production coordinator sitting on top of multiple generation steps, with the creator steering tone and revisions instead of rebuilding each scene from scratch.
The same pattern shows up in sponsored residency work. In carolletta's Steven Bakes a Cake post, the creator said Agent One handled story beats, character animation, voiceover, sound design, and the final edit, while carolletta's link post pointed viewers to InVideo directly.
Pricing and availability
The clearest official availability note is on InVideo's pricing page, which says all paid plans include access to 200-plus image, video, audio, and music models, plus an InVideo v4 agent that can create up to 30 minutes of video from a single prompt. The same page says credits function as the platform currency, model and agent prices can change, and on-demand top-ups are available.
That fills in a missing practical detail from the tweets. The creator posts show what Agent One can produce, but the official docs place it inside a broader InVideo stack that mixes first-party agent workflow with model access, stock libraries, and credit-based billing.
A useful outside read is Kingy AI's review, which argues the product's actual novelty is persistent project context. That lines up with both the tweet demos and InVideo's own docs: the system wants a treatment, remembers the brief, and keeps editing inside the same thread instead of resetting the project each time.