LTX 2.5 Pro (Text-to-Video)

LTX 2.5 Pro turns a prompt into a 6 to 10 second 720p or 1080p video with synchronized sound, tuned for final high-fidelity output, from $0.12 a second.

LTX 2.5 Pro is the quality-optimized mode of LTX-2.5, Lightricks' open-source audio-video model: one pass makes the video and its synchronized sound from a text prompt, for final, high-fidelity output, in 16:9 or 9:16, at 720p or 1080p.

Use it for the clips you keep: ads, trailers, social posts and scenes with dialogue and ambience; for quick drafts and longer or higher-resolution clips use LTX 2.5 Fast (Text-to-Video). camera_motion adds one camera move: dolly in, out, left or right, jib up or down, static or a focus shift. fps is 24, 25 or 50.

Inputs: prompt (required, up to 5,000 characters), duration (6, 8 or 10 seconds; 6 by default), resolution (720p or 1080p; 1080p by default), aspect_ratio (16:9 or 9:16), fps (24, 25 or 50; 25 by default), generate_audio (on by default) and camera_motion. Output: video, one MP4 clip.

Price per second of video, sound included: $0.12 at 720p and $0.17 at 1080p, the same with or without sound and at any frame rate. The default 6-second 1080p clip costs $1.02, 10 seconds at 1080p $1.70. You always choose the length (the model's own automatic length is not offered), so the price is known before the run. It works from text: to animate a picture use LTX 2.5 Pro (Image-to-Video). It does not render 1440p or 4K, does not make clips over 10 seconds, does not edit or extend a video and does not take reference pictures or videos.

Price: From $0.12 per second of duration, depending on resolution. One run with default settings costs $1.02.

Run LTX 2.5 Pro (Text-to-Video) via API

curl -X POST https://api.flymy.ai/api/v1/flymyai/ltx-2_5-pro-text-to-video/predict \
  -H "X-API-KEY: $FLYMYAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"A paper boat floats across a calm pond, gentle camera movement","duration":6,"resolution":"720p"}'

API documentation · Create a free account · Pricing

Related models