Skip to main content
Participant
October 16, 2026
Question

working on a video, it is not folling the prompts at all

Use a cinematic Transformers-style mechanical transformation with intricate, physically connected moving parts. The robot crouches and folds into the black Corvette through a rapid, coordinated sequence of rotating joints, retracting limbs, sliding armor plates, and interlocking automotive panels. Shoulder wheels rotate outward and settle into the wheel wells. The chest separates into rigid sections that assemble into the Corvette’s hood and front fascia. Arms fold into the chassis, legs compress into the rear body, and the door panels pivot into the raised vertical-door positions shown in the final image.

Give every movement a powerful sense of weight, momentum, and mechanical precision. Keep individual components solid and recognizable throughout. Show the complete assembly clearly, without melting, rubbery distortion, dissolving, disappearing parts, or smoke hiding the transformation. Finish as the exact left-facing Corvette in the supplied final frame.

    2 replies

    Participant
    October 20, 2026

    Thanks, for the suggestions. I understand that keeping prompts shorter and avoiding trademarked names may help. However, my main issue is that the video title and concept are working, but Firefly does not follow the transformation sequence correctly. I want the mechanical parts to move in a logical order, with the wheels, arms, legs, and body panels remaining solid and connected throughout the transformation. Has anyone found a reliable way to improve the consistency of complex mechanical transformations in Firefly video?

    I want to be able to work on a video for this sample post : Jackspania racing reviews. In the future, I may make video for racing reviews.

    Translate
    Gabriel Grundler Y.
    Community Manager
    Gabriel Grundler
    October 16, 2026

    Hey, defaultjh37k09kahnv

    I'll share some tips with you:


    • Keep it simple: Firefly works best with short, easy-to-read prompts. Very long or complicated prompts (like those generated by ChatGPT) can confuse the model.

    • Watch your video references: For video generation, avoid dark or busy backgrounds. The AI needs a clear contrast to distinguish the main characters from the environment!

    • Copyright & Public Figures: To keep everything commercially safe, prompts requesting trademarked content, celebrities, or political figures will be blocked.

    • Safety Filters (Minors & NSFW): Our safety guidelines are strictly enforced to protect minors. Firefly will not generate images of children or anything child-like. Prompts containing words like boy, girl, kid, or child will trigger the filter. Additionally, any prompts involving nudity are strictly prohibited.



    So, most likely the AI is not following the full prompt because there are words like 'Corvette', 'Transformers', and that's copyright blocked. Try to change those, avoid long and complicated AI prompts, and you'll be set.


    Also, here are some useful links:


    Generate images from text descriptions
    Writing Effective Prompts
    Adobe Firefly Resources


    Best,


    Gabriel

    Translate