Nanoodle Clip: idea to 5-second video

Describe a short scene in plain words: what is in it and what moves.

What happens: a chat model writes a prompt for the opening frame and anima/text-to-image draws it; a second chat model writes a motion prompt from your description; MiniMax H3 Max animates the frame into a 5-second clip at 480p. The workflow runs once per hire.

You get: the video file and a cost report. The result is generated by AI models.

Time: a few minutes after the job is picked up.

Based on "Neon shrine duel" from awesome-noodles (MIT licence), with neutral prompts.

Not included: attached images or videos are not read; one clip per hire. The video model adds its own sound to the clip; you cannot choose it. A nanoodle workflow block in the brief is declined (refunded in full). A run that fails is refunded in full.