Skip to main content
Create short video clips from text prompts, animate still images, or produce talking-head videos with lipsync. All results are saved to your Pixa workspace.

​
Text-to-video

​
CLI

Generate a video from a text description: 
pixa run prompt "a slow pan across a sunlit kitchen counter with coffee steam rising" \
  --model kling-v2.6-pro \
  --aspect-ratio 16:9 --json

​
MCP

Use the generate_media tool with media_type: "video": 
Generate a 5-second product reveal video of sneakers rotating on a white background.

​
Image-to-video

Attach a start frame and describe the motion you want: 
pixa run prompt "slow zoom in with subtle background motion" \
  --attachment start-frame.png --json
The model uses your image as the opening frame and animates from there.

​
Talking video

Attach a face image and provide a script in the prompt to generate a talking-head video: 
pixa run prompt "Hi everyone, welcome to our product launch. Today we are introducing something special." \
  --attachment face.png --json
The model animates the face to speak the provided text with natural lip movement and expression.

​
Lipsync

Attach both a face image and an audio file to sync lip movement to existing audio: 
pixa run prompt "lipsync this face to the audio" \
  --attachment face.png \
  --attachment voiceover.mp3 --json
The --attachment flag is repeatable. For lipsync, provide the face image and audio file as separate attachments.

​
Video settings

​
CLI flags

FlagDescriptionExample
--modelVideo model IDkling-v2.6-pro
--aspect-ratioOutput aspect ratio16:9, 9:16, 1:1
--output-formatFile formatmp4, webm
--num-variationsNumber of variations1-4

​
MCP parameters

The generate_media tool accepts: prompt, model, aspect_ratio, media_type ("video"), output_format, num_variations, and attachments.

​
Batch video generation

Generate multiple video variations: 
pixa run prompt "product reveal animation" \
  --attachment product.png \
  --num-variations 3 --json
To produce the same video in different aspect ratios for different platforms, run separate commands: 
# Landscape for YouTube
pixa run prompt "product demo video" \
  --attachment product.png --aspect-ratio 16:9 --json

# Portrait for TikTok / Reels
pixa run prompt "product demo video" \
  --attachment product.png --aspect-ratio 9:16 --json

​
Tips

For talking videos, write the script as natural speech. Avoid stage directions or formatting — just write what the person should say.
  • Start frame matters. For image-to-video, the quality and composition of your start frame directly affects the output.
  • Keep prompts focused. Describe one clear motion or scene rather than a complex sequence.
  • Use --json in agent workflows so the next step can parse the result from the output.
  • Async jobs. Video generation can take several minutes. Use pixa jobs follow <id> to poll, or in MCP use get_job_status with sync: true.