lightricks/ltx-video-0.9.7 – Run with an API on Replicate

lightricks / ltx-video-0.9.7

DiT-based 13b video generation model, creating 30fps video

Cold

Public
939 runs
L40S
GitHub
Weights
License

Iterate in playground

Run with an API

Playground API Examples README Versions

Input

prompt

*string

Shift + Return to add a new line

A woman with blood on her face and a white tank top looks down and to her right, then back up as she speaks. She has dark hair pulled back, light skin, and her face and chest are covered in blood. The camera angle is a close-up, focused on the woman's face and upper torso. The lighting is dim and blue-toned, creating a somber and intense atmosphere. The scene appears to be from a movie or TV show.A woman with blood on her face and a white tank top looks down and to her right, then back up as she speaks. She has dark hair pulled back, light skin, and her face and chest are covered in blood. The camera angle is a close-up, focused on the woman's face and upper torso. The lighting is dim and blue-toned, creating a somber and intense atmosphere. The scene appears to be from a movie or TV show.

Text prompt for video generation

image

file

Input image for image-to-video generation. If not provided, text-to-video generation will be used.

negative_prompt

string

Shift + Return to add a new line

worst quality, inconsistent motion, blurry, jittery, distortedworst quality, inconsistent motion, blurry, jittery, distorted

Negative prompt for video generation.

Default: "worst quality, inconsistent motion, blurry, jittery, distorted"

width

integer

Width of the output video. Actual width will be a multiple of 32.

Default: 704

height

integer

Height of the output video. Actual height will be a multiple of 32.

Default: 480

num_frames

integer

Number of frames to generate. Actual frame count will be 8N+1 (e.g., 9, 17, 25, 161).

Default: 161

num_inference_steps

integer

Number of denoising steps.

Default: 50

guidance_scale

number

(minimum: 1, maximum: 10)

Guidance scale. Recommended range: 3.0-3.5.

Default: 3

fps

integer

Frames per second for the output video.

Default: 24

seed

integer

Random seed for reproducible results. Leave blank for a random seed.

Run this model in Node.js with one line of code:

npx create-replicate --model=lightricks/ltx-video-0.9.7

or set up a project from scratch

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run lightricks/ltx-video-0.9.7 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "lightricks/ltx-video-0.9.7:b1a80c6dbce390c23bb52aecebc0e09d445ac12136dd4dc539350c76030fc815",
  {
    input: {
      fps: 24,
      width: 704,
      height: 480,
      prompt: "A woman with blood on her face and a white tank top looks down and to her right, then back up as she speaks. She has dark hair pulled back, light skin, and her face and chest are covered in blood. The camera angle is a close-up, focused on the woman's face and upper torso. The lighting is dim and blue-toned, creating a somber and intense atmosphere. The scene appears to be from a movie or TV show.",
      num_frames: 161,
      guidance_scale: 3,
      negative_prompt: "worst quality, inconsistent motion, blurry, jittery, distorted",
      num_inference_steps: 50
    }
  }
);

// To access the file URL:
console.log(output.url()); //=> "http://example.com"

// To write the file to disk:
fs.writeFile("my-image.png", output);

To learn more, take a look at the guide on getting started with Node.js.

Install Replicate’s Python client library:

pip install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import the client:

import replicate

Run lightricks/ltx-video-0.9.7 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "lightricks/ltx-video-0.9.7:b1a80c6dbce390c23bb52aecebc0e09d445ac12136dd4dc539350c76030fc815",
    input={
        "fps": 24,
        "width": 704,
        "height": 480,
        "prompt": "A woman with blood on her face and a white tank top looks down and to her right, then back up as she speaks. She has dark hair pulled back, light skin, and her face and chest are covered in blood. The camera angle is a close-up, focused on the woman's face and upper torso. The lighting is dim and blue-toned, creating a somber and intense atmosphere. The scene appears to be from a movie or TV show.",
        "num_frames": 161,
        "guidance_scale": 3,
        "negative_prompt": "worst quality, inconsistent motion, blurry, jittery, distorted",
        "num_inference_steps": 50
    }
)
print(output)

To learn more, take a look at the guide on getting started with Python.

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Run lightricks/ltx-video-0.9.7 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "lightricks/ltx-video-0.9.7:b1a80c6dbce390c23bb52aecebc0e09d445ac12136dd4dc539350c76030fc815",
    "input": {
      "fps": 24,
      "width": 704,
      "height": 480,
      "prompt": "A woman with blood on her face and a white tank top looks down and to her right, then back up as she speaks. She has dark hair pulled back, light skin, and her face and chest are covered in blood. The camera angle is a close-up, focused on the woman\'s face and upper torso. The lighting is dim and blue-toned, creating a somber and intense atmosphere. The scene appears to be from a movie or TV show.",
      "num_frames": 161,
      "guidance_scale": 3,
      "negative_prompt": "worst quality, inconsistent motion, blurry, jittery, distorted",
      "num_inference_steps": 50
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

Output

{
  "completed_at": "2025-05-13T19:06:50.797937Z",
  "created_at": "2025-05-13T19:04:34.683000Z",
  "data_removed": false,
  "error": null,
  "id": "x1jn9segfdrma0cpsc48z29tv8",
  "input": {
    "fps": 24,
    "width": 704,
    "height": 480,
    "prompt": "A woman with blood on her face and a white tank top looks down and to her right, then back up as she speaks. She has dark hair pulled back, light skin, and her face and chest are covered in blood. The camera angle is a close-up, focused on the woman's face and upper torso. The lighting is dim and blue-toned, creating a somber and intense atmosphere. The scene appears to be from a movie or TV show.",
    "num_frames": 161,
    "guidance_scale": 3,
    "negative_prompt": "worst quality, inconsistent motion, blurry, jittery, distorted",
    "num_inference_steps": 50
  },
  "logs": "Using seed: 10311\nOriginal inputs: width=704, height=480, num_frames=161\nProcessed inputs: width=704, height=480, num_frames=161\n[~] Using Text-to-Video pipeline\n[~] Generating video with prompt: 'A woman with blood on her face and a white tank top looks down and to her right, then back up as she speaks. She has dark hair pulled back, light skin, and her face and chest are covered in blood. The camera angle is a close-up, focused on the woman's face and upper torso. The lighting is dim and blue-toned, creating a somber and intense atmosphere. The scene appears to be from a movie or TV show.'\n  0%|          | 0/50 [00:00<?, ?it/s]\n  2%|▏         | 1/50 [00:00<00:36,  1.34it/s]\n  4%|▍         | 2/50 [00:01<00:26,  1.80it/s]\n  6%|▌         | 3/50 [00:01<00:29,  1.59it/s]\n  8%|▊         | 4/50 [00:02<00:30,  1.51it/s]\n 10%|█         | 5/50 [00:03<00:30,  1.47it/s]\n 12%|█▏        | 6/50 [00:04<00:30,  1.45it/s]\n 14%|█▍        | 7/50 [00:04<00:30,  1.43it/s]\n 16%|█▌        | 8/50 [00:05<00:29,  1.42it/s]\n 18%|█▊        | 9/50 [00:06<00:29,  1.41it/s]\n 20%|██        | 10/50 [00:06<00:28,  1.41it/s]\n 22%|██▏       | 11/50 [00:07<00:27,  1.41it/s]\n 24%|██▍       | 12/50 [00:08<00:27,  1.40it/s]\n 26%|██▌       | 13/50 [00:09<00:26,  1.40it/s]\n 28%|██▊       | 14/50 [00:09<00:25,  1.40it/s]\n 30%|███       | 15/50 [00:10<00:25,  1.40it/s]\n 32%|███▏      | 16/50 [00:11<00:24,  1.40it/s]\n 34%|███▍      | 17/50 [00:11<00:23,  1.40it/s]\n 36%|███▌      | 18/50 [00:12<00:22,  1.40it/s]\n 38%|███▊      | 19/50 [00:13<00:22,  1.39it/s]\n 40%|████      | 20/50 [00:14<00:21,  1.39it/s]\n 42%|████▏     | 21/50 [00:14<00:20,  1.39it/s]\n 44%|████▍     | 22/50 [00:15<00:20,  1.39it/s]\n 46%|████▌     | 23/50 [00:16<00:19,  1.39it/s]\n 48%|████▊     | 24/50 [00:16<00:18,  1.39it/s]\n 50%|█████     | 25/50 [00:17<00:17,  1.39it/s]\n 52%|█████▏    | 26/50 [00:18<00:17,  1.39it/s]\n 54%|█████▍    | 27/50 [00:19<00:16,  1.39it/s]\n 56%|█████▌    | 28/50 [00:19<00:15,  1.39it/s]\n 58%|█████▊    | 29/50 [00:20<00:15,  1.39it/s]\n 60%|██████    | 30/50 [00:21<00:14,  1.39it/s]\n 62%|██████▏   | 31/50 [00:21<00:13,  1.39it/s]\n 64%|██████▍   | 32/50 [00:22<00:12,  1.39it/s]\n 66%|██████▌   | 33/50 [00:23<00:12,  1.39it/s]\n 68%|██████▊   | 34/50 [00:24<00:11,  1.39it/s]\n 70%|███████   | 35/50 [00:24<00:10,  1.39it/s]\n 72%|███████▏  | 36/50 [00:25<00:10,  1.39it/s]\n 74%|███████▍  | 37/50 [00:26<00:09,  1.39it/s]\n 76%|███████▌  | 38/50 [00:27<00:08,  1.39it/s]\n 78%|███████▊  | 39/50 [00:27<00:07,  1.39it/s]\n 80%|████████  | 40/50 [00:28<00:07,  1.38it/s]\n 82%|████████▏ | 41/50 [00:29<00:06,  1.38it/s]\n 84%|████████▍ | 42/50 [00:29<00:05,  1.38it/s]\n 86%|████████▌ | 43/50 [00:30<00:05,  1.38it/s]\n 88%|████████▊ | 44/50 [00:31<00:04,  1.38it/s]\n 90%|█████████ | 45/50 [00:32<00:03,  1.38it/s]\n 92%|█████████▏| 46/50 [00:32<00:02,  1.38it/s]\n 94%|█████████▍| 47/50 [00:33<00:02,  1.38it/s]\n 96%|█████████▌| 48/50 [00:34<00:01,  1.38it/s]\n 98%|█████████▊| 49/50 [00:34<00:00,  1.38it/s]\n100%|██████████| 50/50 [00:35<00:00,  1.38it/s]\n100%|██████████| 50/50 [00:35<00:00,  1.40it/s]\n[+] Video generation complete: output.mp4",
  "metrics": {
    "predict_time": 46.967726748,
    "total_time": 136.114937
  },
  "output": "https://replicate.delivery/xezq/vRjrsgLsBbosG1TMegwb0ly5a72YtwrejnrZdk1VeosUSpYpA/output.mp4",
  "started_at": "2025-05-13T19:06:03.830211Z",
  "status": "succeeded",
  "urls": {
    "stream": "https://stream.replicate.com/v1/files/bcwr-peutmyepbaz2dztdlreeqx4zgutcixf27aop2ivgbzq4un2ujegq",
    "get": "https://api.replicate.com/v1/predictions/x1jn9segfdrma0cpsc48z29tv8",
    "cancel": "https://api.replicate.com/v1/predictions/x1jn9segfdrma0cpsc48z29tv8/cancel"
  },
  "version": "b1a80c6dbce390c23bb52aecebc0e09d445ac12136dd4dc539350c76030fc815"
}

Generated in

47.0 seconds

Tweak it ShareReport

Using seed: 10311
Original inputs: width=704, height=480, num_frames=161
Processed inputs: width=704, height=480, num_frames=161
[~] Using Text-to-Video pipeline
[~] Generating video with prompt: 'A woman with blood on her face and a white tank top looks down and to her right, then back up as she speaks. She has dark hair pulled back, light skin, and her face and chest are covered in blood. The camera angle is a close-up, focused on the woman's face and upper torso. The lighting is dim and blue-toned, creating a somber and intense atmosphere. The scene appears to be from a movie or TV show.'
  0%|          | 0/50 [00:00<?, ?it/s]
  2%|▏         | 1/50 [00:00<00:36,  1.34it/s]
  4%|▍         | 2/50 [00:01<00:26,  1.80it/s]
  6%|▌         | 3/50 [00:01<00:29,  1.59it/s]
  8%|▊         | 4/50 [00:02<00:30,  1.51it/s]
 10%|█         | 5/50 [00:03<00:30,  1.47it/s]
 12%|█▏        | 6/50 [00:04<00:30,  1.45it/s]
 14%|█▍        | 7/50 [00:04<00:30,  1.43it/s]
 16%|█▌        | 8/50 [00:05<00:29,  1.42it/s]
 18%|█▊        | 9/50 [00:06<00:29,  1.41it/s]
 20%|██        | 10/50 [00:06<00:28,  1.41it/s]
 22%|██▏       | 11/50 [00:07<00:27,  1.41it/s]
 24%|██▍       | 12/50 [00:08<00:27,  1.40it/s]
 26%|██▌       | 13/50 [00:09<00:26,  1.40it/s]
 28%|██▊       | 14/50 [00:09<00:25,  1.40it/s]
 30%|███       | 15/50 [00:10<00:25,  1.40it/s]
 32%|███▏      | 16/50 [00:11<00:24,  1.40it/s]
 34%|███▍      | 17/50 [00:11<00:23,  1.40it/s]
 36%|███▌      | 18/50 [00:12<00:22,  1.40it/s]
 38%|███▊      | 19/50 [00:13<00:22,  1.39it/s]
 40%|████      | 20/50 [00:14<00:21,  1.39it/s]
 42%|████▏     | 21/50 [00:14<00:20,  1.39it/s]
 44%|████▍     | 22/50 [00:15<00:20,  1.39it/s]
 46%|████▌     | 23/50 [00:16<00:19,  1.39it/s]
 48%|████▊     | 24/50 [00:16<00:18,  1.39it/s]
 50%|█████     | 25/50 [00:17<00:17,  1.39it/s]
 52%|█████▏    | 26/50 [00:18<00:17,  1.39it/s]
 54%|█████▍    | 27/50 [00:19<00:16,  1.39it/s]
 56%|█████▌    | 28/50 [00:19<00:15,  1.39it/s]
 58%|█████▊    | 29/50 [00:20<00:15,  1.39it/s]
 60%|██████    | 30/50 [00:21<00:14,  1.39it/s]
 62%|██████▏   | 31/50 [00:21<00:13,  1.39it/s]
 64%|██████▍   | 32/50 [00:22<00:12,  1.39it/s]
 66%|██████▌   | 33/50 [00:23<00:12,  1.39it/s]
 68%|██████▊   | 34/50 [00:24<00:11,  1.39it/s]
 70%|███████   | 35/50 [00:24<00:10,  1.39it/s]
 72%|███████▏  | 36/50 [00:25<00:10,  1.39it/s]
 74%|███████▍  | 37/50 [00:26<00:09,  1.39it/s]
 76%|███████▌  | 38/50 [00:27<00:08,  1.39it/s]
 78%|███████▊  | 39/50 [00:27<00:07,  1.39it/s]
 80%|████████  | 40/50 [00:28<00:07,  1.38it/s]
 82%|████████▏ | 41/50 [00:29<00:06,  1.38it/s]
 84%|████████▍ | 42/50 [00:29<00:05,  1.38it/s]
 86%|████████▌ | 43/50 [00:30<00:05,  1.38it/s]
 88%|████████▊ | 44/50 [00:31<00:04,  1.38it/s]
 90%|█████████ | 45/50 [00:32<00:03,  1.38it/s]
 92%|█████████▏| 46/50 [00:32<00:02,  1.38it/s]
 94%|█████████▍| 47/50 [00:33<00:02,  1.38it/s]
 96%|█████████▌| 48/50 [00:34<00:01,  1.38it/s]
 98%|█████████▊| 49/50 [00:34<00:00,  1.38it/s]
100%|██████████| 50/50 [00:35<00:00,  1.38it/s]
100%|██████████| 50/50 [00:35<00:00,  1.40it/s]
[+] Video generation complete: output.mp4

Examples

View more examples

Run time and cost

This model costs approximately $0.13 to run on Replicate, or 7 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia L40S GPU hardware. Predictions typically complete within 130 seconds.

Readme

LTX-Video-0.9.7 is the latest high-performance video generation model from Lightricks built on DiT. It can generate smooth, high-quality videos in real time—faster than you can watch them. At 1216×704 resolution and 30 frames per second, it produces realistic, dynamic scenes using text, images, or both.

You can control nearly every part of the output. Use a text prompt, or start with an image to animate it. Adjust the width, height, and number of frames to fit your needs. Fine-tune the motion and look of the video using parameters like negative prompt, guidance scale, and denoising steps. You can also set a custom frame rate and seed for repeatable results.