Skip to main content
G
ggblk
Participant
May 15, 2026
Answered

Firefly producing results completely different than I asked for

  • May 15, 2026
  • 1 reply
  • 12 views

I’m new to Firefly so I expect a learning curve, but what I can’t get my head around or fix is Firefly giving me results nowhere near what I asked for.  For example, I told it to add moving film grain to a still image, which it did, but added a decorative frame and did a zoom in/out.   Next I told it to produce the same effect but no frame and no zoom.  It game me the image with no grain effect and tilted away from the camera at an angle.

Is this simply a glitch or am making prompt mistakes?

    Correct answer Hajar Toutsi

    Hi ggblk,
    thanks for reaching out and sharing your experience.

    I understand how confusing it can be when the results don’t match what you had in mind, but that’s part of the learning curve with generative AI.
    Firefly interprets prompts creatively, so the more precise and detailed you are, the closer the output will be to your vision.
     

    Here are a few tips that often help:

    • Be explicit: Don’t assume the AI will preserve elements automatically. Spell out what should stay unchanged.

    • Avoid negative prompting: Instead of saying “no frame, no zoom,” describe exactly what you do want.

    • Choose the right model: Some models handle visual effects better than others.

    • Use detailed phrasing: For example, “Keep all elements of the image intact and add a subtle moving film grain effect.”
       

    If the output still feels off, you can submit feedback directly by clicking the thumbs‑down icon when hovering over the result. That feedback goes straight to the team to help improve Firefly’s behavior.
     

    If you’d like, share the exact prompt and reference image you used, and we can help refine it together.

    Hajar

    1 reply

    Hajar Toutsi
    Community Manager
    Hajar ToutsiCommunity ManagerCorrect answer
    Community Manager
    May 15, 2026

    Hi ggblk,
    thanks for reaching out and sharing your experience.

    I understand how confusing it can be when the results don’t match what you had in mind, but that’s part of the learning curve with generative AI.
    Firefly interprets prompts creatively, so the more precise and detailed you are, the closer the output will be to your vision.
     

    Here are a few tips that often help:

    • Be explicit: Don’t assume the AI will preserve elements automatically. Spell out what should stay unchanged.

    • Avoid negative prompting: Instead of saying “no frame, no zoom,” describe exactly what you do want.

    • Choose the right model: Some models handle visual effects better than others.

    • Use detailed phrasing: For example, “Keep all elements of the image intact and add a subtle moving film grain effect.”
       

    If the output still feels off, you can submit feedback directly by clicking the thumbs‑down icon when hovering over the result. That feedback goes straight to the team to help improve Firefly’s behavior.
     

    If you’d like, share the exact prompt and reference image you used, and we can help refine it together.

    Hajar

    Terms & ConditionsAccessibility statement
    Using the community Terms of use Privacy Cookie preferences Do not sell or share my personal information ad choicesAdChoices