How can we help?

The LTX API gives you text-to-video, image-to-video, and audio-to-video generation — plus retake (re-generate a section of an existing clip) and extend (add frames to the start or end). Every output includes synchronized audio. It's a REST API, so you can drop it into any stack.

Fast is optimised for speed and cost — good for prototyping and iteration, supports clips up to 20 seconds. Pro delivers higher fidelity and motion stability, and is required for audio-to-video, retake, and extend. A good workflow: use Fast to explore, switch to Pro for final renders. See the full breakdown on the supported models page.

The sync API returns the video in a single HTTP call — simplest for short clips and quick experiments. The async API submits a job and lets you poll for the result, which is better for production where you don't want to hold a long-lived connection open. Both support the same endpoints. See async jobs for details.

You're billed per second of output video. The rate depends on which model you use (Fast or Pro) and the resolution (1080p, 1440p, or 4K). No minimums, no hidden fees. Full rate card on the pricing page — enterprise plans available via sales.

Sign up at the developer console, generate an API key, and follow the quickstart guide to make your first request in minutes. Authentication uses a Bearer token — include your key in the Authorization header of every request.

By default, you can run 2 generations concurrently. Exceeding this returns a 429 with a Retry-After header. If you need higher throughput, contact our team — enterprise limits are available. Full details in the rate limits docs.