Generate a 15-second video from an image using Grok Imagine 1.5
Free path via Figma Weave: upload an image, write a prompt - get a finished clip with camera control and a stable character
- 1.Open weave.figma.com and sign up. You get 150 free credits for the month upon registration - enough for 15 videos at 10 credits each.
If the service is unavailable in your region, try using a VPN. Credits refresh monthly for each account.
Figma Weave ↗ - 2.Open the model page on fal.ai in a new tab - fal.ai/models/xai/grok-imagine-video/v1.5/image-to-video/. Copy the URL from the address bar - you'll need it to import the model into Weave.
You don't need to register on fal.ai if you're working through Weave - it uses its own credits.
FAL ↗About this tool - 3.Go back to Figma Weave. Right-click on the empty canvas and select «Import Model». Paste the copied Grok Imagine 1.5 model URL. The model will appear on the canvas with a settings panel on the right.
If the import didn't work, make sure you copied the full URL including /image-to-video/ at the end.
Figma Weave ↗ - 4.Configure the model settings on the right: duration - 15 seconds, resolution - 720. The generation cost will show as 10 credits. Right-click the canvas and add an Image Input node. Upload your reference image - this will be the starting frame of the video.
Images with a clear character or object in the center and a neutral background work best.
Figma Weave ↗ - 5.Connect the output of the image node to the model's Image input. Then add a Prompt node (right-click -> Prompt), connect it to the model's Prompt input, and paste the following text, replacing the content in square brackets with your own:
A prompt in English gives more predictable results. Describe the camera movement, the character's action, and the environment.
Figma Weave ↗Prompt[Character description, e.g. a robot in a trench coat] walks slowly through [setting, e.g. an abandoned neon-lit city street]. Camera pulls back to reveal the scale of the environment. Cinematic lighting, dramatic atmosphere, smooth camera movement.
What this prompt doesThis prompt describes the movement of the character and the camera. Replace [Character description...] with your character's description, and [setting...] with your environment description. «Camera pulls back» means the camera moves backward, revealing the scale of the scene. - 6.Click the «Run» button. Generation takes 1 - 3 minutes. The finished video will appear in the output node - download it using the Download button.
If you don't like the result - change the prompt and run it again. Each attempt costs 10 credits; you have 150 per month.
Figma Weave ↗
- freeFigma Weave - free plan150 credits monthly upon registration; one 15-second 720p video costs 10 credits - so 15 videos per month for free.
Why today
AI services change fast - interfaces and free limits may differ from what's described.