Flagship video, without the flagship bill
The quality of the best models, priced fairly, on infrastructure built to stay up.
Lower prices
Pay per request — no subscription, no minimums, and credits never expire. Flagship models at rates indie teams can actually afford.
Higher quality
Veo 3.1, Omni Flash — cinematic motion and native sound, with no quality downgrade.
Stable routing
Multi-route smart scheduling. If one upstream stalls, your job reroutes automatically — generations keep flowing.
Generate and edit video in one place
From a single prompt to a finished, edited clip — no GPUs, no pipeline glue.
Text to video
Describe a shot, get a cinematic 1080p/4K clip with sound. Veo 3.1, Omni Flash.
Image to video
Bring any still to life — camera moves, motion, and consistent characters from a single frame.
AI video editing
Restyle, extend, upscale, interpolate frames and transfer motion — refine your clip without leaving the canvas.
Every top video model, one endpoint
Veo 3.1, Omni Flash and more — swap with one line of code, no lock-in.
Real clips, generated on GenRelay
Text-to-video, image-to-video, and edits — straight from the platform.
Cinematic clips from a single line
Prompt to 1080p/4K video with native sound — Veo 3.1, Omni Flash.
Animate any still
Camera moves and motion from a single frame.
Match a reference clip
Feed one clip; new shots come back with its look, pacing and camera.
Restyle the whole take
Recast a clip in a new material and palette, frame for frame.
Integrate in minutes
A clean, consistent API across every model. Generate your first asset with three lines of code.
- Unified REST & SDK (Python, JS, Go)
- Real-time webhooks & streaming
- Usage analytics & per-key spend limits
- Private endpoints
# pip install genrelay from genrelay import GenRelay client = GenRelay(api_key="sk-...") video = client.videos.generate( model="veo_3_1", prompt="a drone shot over neon Tokyo", seconds=8, tier="1080p", ) print(video.output_url)