DeepInfra raises $107M Series B to scale the inference cloud — read the announcement
Lightricks/
$0.17 / second
*LTX-2.5 Pro is Lightricks' higher-fidelity LTX-2.5 model. It generates video from text or a first frame in 16:9, 9:16, 1:1, 3:2 or 2:3, up to 4K at 24–50 fps, with synchronized audio, native multi-shot scenes and clips of 6, 8 or 10 seconds.

Prompt
Text prompt describing the video, or how to animate the image
You need to log in to use this model
Log InSettings
Please upload an image file
Please upload an image file
Resolution
Output resolution as WIDTHxHEIGHT, in 16:9, 9:16, 1:1, 3:2 or 2:3. Price per second by the short side: 720p $0.12, 1080p $0.17, 1440p $0.25, 4K $0.39
Duration
Duration in seconds
Fps
Frame rate in frames per second
Generate Audio
Whether to generate an audio track matching the scene
Camera Motion
Camera motion effect applied to the video
LTX-2.5 Pro is the higher-fidelity variant of LTX-2.5, Lightricks' latest video generation model. It turns a text prompt or a starting image into video up to 4K, in landscape, portrait or square, at cinematic frame rates, with an AI-generated audio track that matches the scene.
LTX-2.5 improves every stage of generation: a diffusion video decoder for higher pixel quality, finer detail in texture, faces and typography, stronger prompt adherence, and native multi-shot scenes. A single generation can produce several connected shots that keep character, scene, lighting, visual style and voice consistent across cuts.
generate_audio: false| 16:9 | 9:16 | 1:1 | 3:2 | 2:3 | |
|---|---|---|---|---|---|
| 720p | 1280x720 | 720x1280 | 720x720 | 1080x720 | 720x1080 |
| 1080p | 1920x1080 | 1080x1920 | 1080x1080 | 1620x1080 | 1080x1620 |
| 1440p | 2560x1440 | 1440x2560 | 1440x1440 | 2160x1440 | 1440x2160 |
| 4K | 3840x2160 | 2160x3840 | 2160x2160 | 3240x2160 | 2160x3240 |
6, 8 or 10 seconds at every resolution and at 24, 25, 48 or 50 fps.
Pass image_url to animate an image from its first frame, and optionally last_frame_url to set where the clip ends. Each takes an https URL (up to 15 MB), a base64 data URI or a file upload (up to 7 MB encoded), in PNG, JPEG or WebP. Image-to-video costs the same as text-to-video.
Billed per second of generated video, by resolution:
| Resolution | Price per second |
|---|---|
| 720p | $0.12 |
| 1080p | $0.17 |
| 1440p | $0.25 |
| 4K | $0.39 |
A 6-second 720p clip costs $0.72. A 10-second 4K clip costs $3.90.
© 2026 DeepInfra. All rights reserved.