If your video generations aren't matching what you're after, try these in order.
1. Regenerate
AI video generations vary run-to-run. Re-running the same prompt or template often produces a better result on the second or third try.
2. Tweak your prompt
If you're using Custom prompt (Text-to-Video) or Image-to-Video:
Describe one main action clearly — "turning her head to smile at the camera" beats "happy"
Keep prompts visual — describe what we'd see, not what the character is feeling
Pick the moment you want to capture, since Foxy generates short clips
If you started from a template: the prompt is locked. To customize, start from Create → Custom prompt instead.
3. Try Image-to-Video
If you have a great image you want to bring to life, Image-to-Video is more likely to preserve the look you want than Text-to-Video from scratch. Open the image in your gallery → tap the play icon → write a short action prompt.
4. Check your character
If the character doesn't look like you in videos, the issue is your training photos. See My AI doesn't look like me — what can I do? for how to retrain (free, no new slot needed).
Still stuck?
Open chat with us via the Profile menu → Support with the video and prompt — we can suggest specifics.
