Video generation in Layer uses third-party model providers. Most issues fall into one of a handful of categories. This article covers how to identify what is happening and what to do about it.
Common issues at a glance
Issue | Most likely cause |
Generation fails with a generic error | Model provider instability or a session state issue |
Generation runs a long time without completing | Provider slowness during high demand (can be especially common during highly anticipated new model releases); source video too short for the model's requirements |
Safety or NSFW rejection on non-explicit content | AI content safety false positive; model-specific sensitivity |
Output video crops or misframes the reference image | Aspect ratio mismatch between your reference image and the selected output ratio |
First or Last Frame option not available | Not every video model supports both reference frames — check the model's page in Base Models |
Reframe producing incorrect or cropped output | Low source resolution, or a content safety trigger during reframing |
Generation fails with an error
If a video generation returns an error immediately or partway through, the most likely cause is provider-side instability. Check whether the issue is isolated to one specific model or affecting all video models. If only one model is affected, this almost always points to the provider rather than Layer.
What to try:
Retry the generation once. Some errors are transient and a second attempt on the same model will succeed.
Switch to an alternative video model — open Base Models and filter by Video, or just ask the Agent to try a different one.
Check the Layer Status Page for any active incidents.
If the error persists across multiple models, start a fresh session from Home and try again.
Note that failed video generations do not deduct Creative Units. You are only charged when a generation completes successfully.
Generation is very slow or appears stuck
Video generations take longer than image generations by nature, and some models slow down further during periods of high demand.
What to try:
Check the model's typical generation time on its Base Models page, and try a faster alternative if speed matters more than maximum quality for this generation.
If a generation has been running for an unusually long time and appears frozen, refresh the page — it may have completed in the background and will show on your board or in the Library.
If you're animating from a source video (not just an image), check that it meets the minimum length the model requires — very short clips can cause a failure rather than a slow generation.
Safety or NSFW rejection on non-explicit content
Video models use AI-based content safety checks that can produce false positives on completely non-explicit content — gaming assets, character animations, and hand or object movement have all been flagged incorrectly in practice. This is a behaviour of the model provider's safety system, not something Layer can override directly.
What to try:
Retry the generation with the same prompt and model — the safety check runs fresh each time and the same content may pass on a subsequent attempt.
Simplify the prompt. Very detailed or descriptive prompts occasionally trigger safety systems more than a shorter, more neutral version of the same instruction.
Switch to a different model — each provider has its own content safety policy, so a rejection on one model doesn't mean every model will reject it.
Output video crops the reference image
If the generated video only shows a portion of your reference image, or the subject appears cropped, this is almost always caused by a mismatch between your reference image's aspect ratio and the aspect ratio selected for the video output.
What to try:
Match your reference image's dimensions to your selected output aspect ratio.
If you need to change the aspect ratio of an already-generated video, use the Reframe tool from the lightbox — open the video, select your target ratio, and write a prompt describing how the frame should be expanded (for example, "expand the sides to fit landscape format").
First Frame and Last Frame support
Not all video models support both first and last frame references — check the model's page on Base Models to confirm. If a workspace admin has disabled a model you need, they can re-enable it from that model's page.
Editing an existing video
To modify a video you've already generated rather than starting fresh, open it in the lightbox and use Prompt edit — the same editing pattern used across every modality in Layer. Describe the change you want (for example, "make the lighting warmer") and Layer applies it as a new generation, leaving your original untouched.
Reframe issues
The Reframe tool changes the aspect ratio or framing of an existing video. If it's producing unexpected output or failing with a content safety rejection:
Check that your source video is reasonably high resolution — lower-resolution sources give the model less to work with and can produce poor results.
If Reframe returns a safety rejection on content that is clearly not explicit, retry once and then try a different model if one is available.
Write a specific reframe prompt describing the desired output — vague or empty prompts can produce inconsistent results.
Frequently Asked Questions
My video failed partway through. Were CUs charged?
No. Creative Units are only deducted when a generation completes successfully. You can confirm this in Settings → Usage.
How long should video generation normally take?
This varies significantly by model and duration — check the typical generation time listed on each model's Base Models page. If a generation exceeds 30 minutes without completing, refresh the page to check your board or Library, then contact support with your session link if it's still not there.
Can I use my trained LoRA with video generation?
Custom-trained LoRA models are built for image generation. Check the model's capabilities on its Base Models page to confirm whether style training carries over to a given video model.
The output video looks nothing like my reference image. What can I do?
First check that your reference image matches your selected output aspect ratio. If the ratios match and the output still drifts from the reference, try a more descriptive prompt with specific visual details, or try a model with stronger First/Last Frame support.
Still experiencing issues with video generation? Contact support with your session link and the specific model and prompt you were using and we will investigate.