lucataco/wan-2.1-1.3b-vid2vid | Run with an API on Replicate

lucataco / wan-2.1-1.3b-vid2vid

Wan 2.1 1.3b Video to Video. Wan is a powerful visual generation model developed by Tongyi Lab of Alibaba Group

Cold

Public
765 runs
L40S
GitHub
Weights
Paper
License

Iterate in playground

Run with an API

Playground API Examples README Versions

Input

prompt

*string

Shift + Return to add a new line

Text prompt describing what you want to generate or modify

negative_prompt

string

Shift + Return to add a new line

low quality, blurry, distorted, disfigured, text, watermarklow quality, blurry, distorted, disfigured, text, watermark

Negative prompt to specify what to avoid in the generation

Default: "low quality, blurry, distorted, disfigured, text, watermark"

input_video

file

Input video for video-to-video generation

num_frames

integer

(minimum: 1, maximum: 100)

Number of frames to generate in the output video

Default: 81

frames_per_second

integer

(minimum: 5, maximum: 24)

Number of frames per second in the output video

Default: 16

denoising_strength

number

(minimum: 0, maximum: 1)

Strength of denoising when using video-to-video mode. Higher values change more content.

Default: 0.7

aspect_ratio

string

Aspect ratio for the output video

Default: "832x480"

num_inference_steps

integer

(minimum: 10, maximum: 50)

Number of sampling steps (higher = better quality but slower)

Default: 40

cfg_scale

number

(minimum: 0, maximum: 20)

Classifier free guidance scale (higher values strengthen prompt adherence)

Default: 6

seed

integer

Random seed for reproducible results (leave blank for random)

tiled

boolean

Whether to use tiled sampling for better quality on larger videos

Default: true

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

npx create-replicate --model=lucataco/wan-2.1-1.3b-vid2vid

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";
import fs from "node:fs";

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

Run lucataco/wan-2.1-1.3b-vid2vid using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "lucataco/wan-2.1-1.3b-vid2vid:9349766527ed95fa6194dcca4cae3d497357e207025beb0b97fb0403420142b8",
  {
    input: {
      tiled: true,
      prompt: "a puppy wearing black sunglasses",
      cfg_scale: 6,
      num_frames: 81,
      input_video: "https://replicate.delivery/pbxt/MbUZEHGJFGoTaCAlbxPCAHvkmQrWOwAWGhb3o6xwdYRtoV7C/replicate-prediction-6jxv92xhadrma0cncbabf652xm.mp4",
      aspect_ratio: "832x480",
      negative_prompt: "low quality, blurry, distorted, disfigured, text, watermark",
      frames_per_second: 16,
      denoising_strength: 0.7,
      num_inference_steps: 30
    }
  }
);

// 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 lucataco/wan-2.1-1.3b-vid2vid using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "lucataco/wan-2.1-1.3b-vid2vid:9349766527ed95fa6194dcca4cae3d497357e207025beb0b97fb0403420142b8",
    input={
        "tiled": True,
        "prompt": "a puppy wearing black sunglasses",
        "cfg_scale": 6,
        "num_frames": 81,
        "input_video": "https://replicate.delivery/pbxt/MbUZEHGJFGoTaCAlbxPCAHvkmQrWOwAWGhb3o6xwdYRtoV7C/replicate-prediction-6jxv92xhadrma0cncbabf652xm.mp4",
        "aspect_ratio": "832x480",
        "negative_prompt": "low quality, blurry, distorted, disfigured, text, watermark",
        "frames_per_second": 16,
        "denoising_strength": 0.7,
        "num_inference_steps": 30
    }
)
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 lucataco/wan-2.1-1.3b-vid2vid 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": "lucataco/wan-2.1-1.3b-vid2vid:9349766527ed95fa6194dcca4cae3d497357e207025beb0b97fb0403420142b8",
    "input": {
      "tiled": true,
      "prompt": "a puppy wearing black sunglasses",
      "cfg_scale": 6,
      "num_frames": 81,
      "input_video": "https://replicate.delivery/pbxt/MbUZEHGJFGoTaCAlbxPCAHvkmQrWOwAWGhb3o6xwdYRtoV7C/replicate-prediction-6jxv92xhadrma0cncbabf652xm.mp4",
      "aspect_ratio": "832x480",
      "negative_prompt": "low quality, blurry, distorted, disfigured, text, watermark",
      "frames_per_second": 16,
      "denoising_strength": 0.7,
      "num_inference_steps": 30
    }
  }' \
  https://api.replicate.com/v1/predictions

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

Output

{
  "completed_at": "2025-04-02T18:03:40.537024Z",
  "created_at": "2025-04-02T18:01:26.681000Z",
  "data_removed": false,
  "error": null,
  "id": "3aybzba7k5rmc0cnyynbneaztg",
  "input": {
    "tiled": true,
    "prompt": "a puppy wearing black sunglasses",
    "cfg_scale": 6,
    "num_frames": 81,
    "input_video": "https://replicate.delivery/pbxt/MbUZEHGJFGoTaCAlbxPCAHvkmQrWOwAWGhb3o6xwdYRtoV7C/replicate-prediction-6jxv92xhadrma0cncbabf652xm.mp4",
    "aspect_ratio": "832x480",
    "negative_prompt": "low quality, blurry, distorted, disfigured, text, watermark",
    "frames_per_second": 16,
    "denoising_strength": 0.7,
    "num_inference_steps": 30
  },
  "logs": "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\nTo disable this warning, you can either:\n- Avoid using `tokenizers` before the fork if possible\n- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\nVAE encoding:   0%|          | 0/9 [00:00<?, ?it/s]\nVAE encoding:  11%|█         | 1/9 [00:00<00:05,  1.58it/s]\nVAE encoding:  22%|██▏       | 2/9 [00:01<00:04,  1.58it/s]\nVAE encoding:  33%|███▎      | 3/9 [00:01<00:03,  1.58it/s]\nVAE encoding:  44%|████▍     | 4/9 [00:02<00:03,  1.57it/s]\nVAE encoding:  56%|█████▌    | 5/9 [00:03<00:02,  1.57it/s]\nVAE encoding:  67%|██████▋   | 6/9 [00:03<00:01,  1.58it/s]\nVAE encoding:  78%|███████▊  | 7/9 [00:04<00:01,  1.57it/s]\nVAE encoding:  89%|████████▉ | 8/9 [00:05<00:00,  1.57it/s]\nVAE encoding: 100%|██████████| 9/9 [00:05<00:00,  1.57it/s]\nVAE encoding: 100%|██████████| 9/9 [00:05<00:00,  1.57it/s]\n  0%|          | 0/30 [00:00<?, ?it/s]\n  3%|▎         | 1/30 [00:03<01:44,  3.60s/it]\n  7%|▋         | 2/30 [00:07<01:40,  3.60s/it]\n 10%|█         | 3/30 [00:10<01:37,  3.60s/it]\n 13%|█▎        | 4/30 [00:14<01:33,  3.60s/it]\n 17%|█▋        | 5/30 [00:17<01:29,  3.60s/it]\n 20%|██        | 6/30 [00:21<01:26,  3.61s/it]\n 23%|██▎       | 7/30 [00:25<01:22,  3.60s/it]\n 27%|██▋       | 8/30 [00:28<01:19,  3.60s/it]\n 30%|███       | 9/30 [00:32<01:15,  3.60s/it]\n 33%|███▎      | 10/30 [00:36<01:12,  3.60s/it]\n 37%|███▋      | 11/30 [00:39<01:08,  3.61s/it]\n 40%|████      | 12/30 [00:43<01:04,  3.60s/it]\n 43%|████▎     | 13/30 [00:46<01:01,  3.60s/it]\n 47%|████▋     | 14/30 [00:50<00:57,  3.60s/it]\n 50%|█████     | 15/30 [00:54<00:54,  3.60s/it]\n 53%|█████▎    | 16/30 [00:57<00:50,  3.60s/it]\n 57%|█████▋    | 17/30 [01:01<00:46,  3.60s/it]\n 60%|██████    | 18/30 [01:04<00:43,  3.60s/it]\n 63%|██████▎   | 19/30 [01:08<00:39,  3.60s/it]\n 67%|██████▋   | 20/30 [01:12<00:35,  3.60s/it]\n 70%|███████   | 21/30 [01:15<00:32,  3.60s/it]\n 73%|███████▎  | 22/30 [01:19<00:28,  3.61s/it]\n 77%|███████▋  | 23/30 [01:22<00:25,  3.61s/it]\n 80%|████████  | 24/30 [01:26<00:21,  3.61s/it]\n 83%|████████▎ | 25/30 [01:30<00:18,  3.61s/it]\n 87%|████████▋ | 26/30 [01:33<00:14,  3.61s/it]\n 90%|█████████ | 27/30 [01:37<00:10,  3.61s/it]\n 93%|█████████▎| 28/30 [01:40<00:07,  3.61s/it]\n 97%|█████████▋| 29/30 [01:44<00:03,  3.61s/it]\n100%|██████████| 30/30 [01:48<00:00,  3.60s/it]\n100%|██████████| 30/30 [01:48<00:00,  3.60s/it]\nVAE decoding:   0%|          | 0/9 [00:00<?, ?it/s]\nVAE decoding:  11%|█         | 1/9 [00:01<00:08,  1.12s/it]\nVAE decoding:  22%|██▏       | 2/9 [00:02<00:07,  1.11s/it]\nVAE decoding:  33%|███▎      | 3/9 [00:03<00:06,  1.11s/it]\nVAE decoding:  44%|████▍     | 4/9 [00:04<00:05,  1.11s/it]\nVAE decoding:  56%|█████▌    | 5/9 [00:05<00:04,  1.11s/it]\nVAE decoding:  67%|██████▋   | 6/9 [00:06<00:03,  1.11s/it]\nVAE decoding:  78%|███████▊  | 7/9 [00:07<00:02,  1.11s/it]\nVAE decoding:  89%|████████▉ | 8/9 [00:08<00:01,  1.11s/it]\nVAE decoding: 100%|██████████| 9/9 [00:09<00:00,  1.11s/it]\nVAE decoding: 100%|██████████| 9/9 [00:09<00:00,  1.11s/it]\nSaving video:   0%|          | 0/81 [00:00<?, ?it/s]huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\nTo disable this warning, you can either:\n- Avoid using `tokenizers` before the fork if possible\n- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\nSaving video:   1%|          | 1/81 [00:00<00:23,  3.33it/s]\nSaving video:  89%|████████▉ | 72/81 [00:00<00:00, 230.25it/s]\nSaving video: 100%|██████████| 81/81 [00:00<00:00, 186.99it/s]",
  "metrics": {
    "predict_time": 133.847287778,
    "total_time": 133.856024
  },
  "output": "https://replicate.delivery/xezq/8pFLomYr6wrQIJOYErDNrnZ5xwqQLiirI1yuB2ZmDSDfbZPKA/tmp_vurc1dr.mp4",
  "started_at": "2025-04-02T18:01:26.689737Z",
  "status": "succeeded",
  "urls": {
    "stream": "https://stream.replicate.com/v1/files/bcwr-fowizvzrn7x5glwymckz5k4gsffnz2kfshm6axaeqkzvqylmvorq",
    "get": "https://api.replicate.com/v1/predictions/3aybzba7k5rmc0cnyynbneaztg",
    "cancel": "https://api.replicate.com/v1/predictions/3aybzba7k5rmc0cnyynbneaztg/cancel"
  },
  "version": "9349766527ed95fa6194dcca4cae3d497357e207025beb0b97fb0403420142b8"
}

Generated in

2 minutes 14 seconds

Tweak it ShareReport View full prediction

huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
VAE encoding:   0%|          | 0/9 [00:00<?, ?it/s]
VAE encoding:  11%|█         | 1/9 [00:00<00:05,  1.58it/s]
VAE encoding:  22%|██▏       | 2/9 [00:01<00:04,  1.58it/s]
VAE encoding:  33%|███▎      | 3/9 [00:01<00:03,  1.58it/s]
VAE encoding:  44%|████▍     | 4/9 [00:02<00:03,  1.57it/s]
VAE encoding:  56%|█████▌    | 5/9 [00:03<00:02,  1.57it/s]
VAE encoding:  67%|██████▋   | 6/9 [00:03<00:01,  1.58it/s]
VAE encoding:  78%|███████▊  | 7/9 [00:04<00:01,  1.57it/s]
VAE encoding:  89%|████████▉ | 8/9 [00:05<00:00,  1.57it/s]
VAE encoding: 100%|██████████| 9/9 [00:05<00:00,  1.57it/s]
VAE encoding: 100%|██████████| 9/9 [00:05<00:00,  1.57it/s]
  0%|          | 0/30 [00:00<?, ?it/s]
  3%|▎         | 1/30 [00:03<01:44,  3.60s/it]
  7%|▋         | 2/30 [00:07<01:40,  3.60s/it]
 10%|█         | 3/30 [00:10<01:37,  3.60s/it]
 13%|█▎        | 4/30 [00:14<01:33,  3.60s/it]
 17%|█▋        | 5/30 [00:17<01:29,  3.60s/it]
 20%|██        | 6/30 [00:21<01:26,  3.61s/it]
 23%|██▎       | 7/30 [00:25<01:22,  3.60s/it]
 27%|██▋       | 8/30 [00:28<01:19,  3.60s/it]
 30%|███       | 9/30 [00:32<01:15,  3.60s/it]
 33%|███▎      | 10/30 [00:36<01:12,  3.60s/it]
 37%|███▋      | 11/30 [00:39<01:08,  3.61s/it]
 40%|████      | 12/30 [00:43<01:04,  3.60s/it]
 43%|████▎     | 13/30 [00:46<01:01,  3.60s/it]
 47%|████▋     | 14/30 [00:50<00:57,  3.60s/it]
 50%|█████     | 15/30 [00:54<00:54,  3.60s/it]
 53%|█████▎    | 16/30 [00:57<00:50,  3.60s/it]
 57%|█████▋    | 17/30 [01:01<00:46,  3.60s/it]
 60%|██████    | 18/30 [01:04<00:43,  3.60s/it]
 63%|██████▎   | 19/30 [01:08<00:39,  3.60s/it]
 67%|██████▋   | 20/30 [01:12<00:35,  3.60s/it]
 70%|███████   | 21/30 [01:15<00:32,  3.60s/it]
 73%|███████▎  | 22/30 [01:19<00:28,  3.61s/it]
 77%|███████▋  | 23/30 [01:22<00:25,  3.61s/it]
 80%|████████  | 24/30 [01:26<00:21,  3.61s/it]
 83%|████████▎ | 25/30 [01:30<00:18,  3.61s/it]
 87%|████████▋ | 26/30 [01:33<00:14,  3.61s/it]
 90%|█████████ | 27/30 [01:37<00:10,  3.61s/it]
 93%|█████████▎| 28/30 [01:40<00:07,  3.61s/it]
 97%|█████████▋| 29/30 [01:44<00:03,  3.61s/it]
100%|██████████| 30/30 [01:48<00:00,  3.60s/it]
100%|██████████| 30/30 [01:48<00:00,  3.60s/it]
VAE decoding:   0%|          | 0/9 [00:00<?, ?it/s]
VAE decoding:  11%|█         | 1/9 [00:01<00:08,  1.12s/it]
VAE decoding:  22%|██▏       | 2/9 [00:02<00:07,  1.11s/it]
VAE decoding:  33%|███▎      | 3/9 [00:03<00:06,  1.11s/it]
VAE decoding:  44%|████▍     | 4/9 [00:04<00:05,  1.11s/it]
VAE decoding:  56%|█████▌    | 5/9 [00:05<00:04,  1.11s/it]
VAE decoding:  67%|██████▋   | 6/9 [00:06<00:03,  1.11s/it]
VAE decoding:  78%|███████▊  | 7/9 [00:07<00:02,  1.11s/it]
VAE decoding:  89%|████████▉ | 8/9 [00:08<00:01,  1.11s/it]
VAE decoding: 100%|██████████| 9/9 [00:09<00:00,  1.11s/it]
VAE decoding: 100%|██████████| 9/9 [00:09<00:00,  1.11s/it]
Saving video:   0%|          | 0/81 [00:00<?, ?it/s]huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Saving video:   1%|          | 1/81 [00:00<00:23,  3.33it/s]
Saving video:  89%|████████▉ | 72/81 [00:00<00:00, 230.25it/s]
Saving video: 100%|██████████| 81/81 [00:00<00:00, 186.99it/s]

Examples

View more examples

Run time and cost

This model costs approximately $0.20 to run on Replicate, or 5 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 4 minutes. The predict time for this model varies significantly based on the inputs.

Readme

About

Cog implementation of modelscope’s video to video process