jichengdu/wan-i2v – Run with an API on Replicate

jichengdu / wan-i2v

Cold

Public
12 runs
A100 (80GB)

Run with an API

Playground API Examples README Versions

Input

image

file

Input image to start generating from

Default: "examples/i2v_input.JPG"

prompt

string

Shift + Return to add a new line

Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline's intricate details and the refreshing atmosphere of the seaside.Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline's intricate details and the refreshing atmosphere of the seaside.

Prompt for video generation

Default: "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline's intricate details and the refreshing atmosphere of the seaside."

num_frames

integer

(minimum: 81, maximum: 100)

Number of video frames. 81 frames give the best results

Default: 81

max_area

string

Shift + Return to add a new line

Maximum area of generated image. The input image will shrink to fit these dimensions

Default: "1280*720"

frames_per_second

integer

(minimum: 5, maximum: 24)

Frames per second

Default: 16

sample_steps

integer

(minimum: 1, maximum: 40)

Number of generation steps

Default: 30

sample_guide_scale

number

(minimum: 0, maximum: 10)

Higher guide scale makes prompt adherence better

Default: 5

sample_shift

number

(minimum: 1, maximum: 10)

Sample shift factor

Default: 5

seed

integer

Random seed. Leave blank for random

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

npx create-replicate --model=jichengdu/wan-i2v

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 jichengdu/wan-i2v using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "jichengdu/wan-i2v:d6a220d6990c8e166d4639a7b8fc9ef744e8e5297af2aef18edde1477140782e",
  {
    input: {
      image: "https://replicate.delivery/pbxt/MfpzR98ZJmwZM7oUcVqeFNW5DUvtTGFDTopDvFC40JdGhZKB/i2v_input.JPG",
      prompt: "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline's intricate details and the refreshing atmosphere of the seaside.",
      max_area: "1280*720",
      num_frames: 81,
      sample_shift: 5,
      sample_steps: 30,
      frames_per_second: 16,
      sample_guide_scale: 5
    }
  }
);
console.log(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 jichengdu/wan-i2v using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "jichengdu/wan-i2v:d6a220d6990c8e166d4639a7b8fc9ef744e8e5297af2aef18edde1477140782e",
    input={
        "image": "https://replicate.delivery/pbxt/MfpzR98ZJmwZM7oUcVqeFNW5DUvtTGFDTopDvFC40JdGhZKB/i2v_input.JPG",
        "prompt": "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline's intricate details and the refreshing atmosphere of the seaside.",
        "max_area": "1280*720",
        "num_frames": 81,
        "sample_shift": 5,
        "sample_steps": 30,
        "frames_per_second": 16,
        "sample_guide_scale": 5
    }
)
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 jichengdu/wan-i2v 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": "d6a220d6990c8e166d4639a7b8fc9ef744e8e5297af2aef18edde1477140782e",
    "input": {
      "image": "https://replicate.delivery/pbxt/MfpzR98ZJmwZM7oUcVqeFNW5DUvtTGFDTopDvFC40JdGhZKB/i2v_input.JPG",
      "prompt": "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline\'s intricate details and the refreshing atmosphere of the seaside.",
      "max_area": "1280*720",
      "num_frames": 81,
      "sample_shift": 5,
      "sample_steps": 30,
      "frames_per_second": 16,
      "sample_guide_scale": 5
    }
  }' \
  https://api.replicate.com/v1/predictions

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

Install Cog

brew install cog

If you don’t have Homebrew, there are other installation options available.

Pull and run jichengdu/wan-i2v using Cog (this will download the full model and run it in your local environment):

cog predict r8.im/jichengdu/wan-i2v@sha256:d6a220d6990c8e166d4639a7b8fc9ef744e8e5297af2aef18edde1477140782e \
  -i 'image="https://replicate.delivery/pbxt/MfpzR98ZJmwZM7oUcVqeFNW5DUvtTGFDTopDvFC40JdGhZKB/i2v_input.JPG"' \
  -i $'prompt="Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline\'s intricate details and the refreshing atmosphere of the seaside."' \
  -i 'max_area="1280*720"' \
  -i 'num_frames=81' \
  -i 'sample_shift=5' \
  -i 'sample_steps=30' \
  -i 'frames_per_second=16' \
  -i 'sample_guide_scale=5'

To learn more, take a look at the Cog documentation.

Pull and run jichengdu/wan-i2v using Docker (this will download the full model and run it in your local environment):

docker run -d -p 5000:5000 --gpus=all r8.im/jichengdu/wan-i2v@sha256:d6a220d6990c8e166d4639a7b8fc9ef744e8e5297af2aef18edde1477140782e
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "image": "https://replicate.delivery/pbxt/MfpzR98ZJmwZM7oUcVqeFNW5DUvtTGFDTopDvFC40JdGhZKB/i2v_input.JPG",
      "prompt": "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline\'s intricate details and the refreshing atmosphere of the seaside.",
      "max_area": "1280*720",
      "num_frames": 81,
      "sample_shift": 5,
      "sample_steps": 30,
      "frames_per_second": 16,
      "sample_guide_scale": 5
    }
  }' \
  http://localhost:5000/predictions

Output

{
  "completed_at": "2025-03-17T07:57:48.847319Z",
  "created_at": "2025-03-17T07:11:31.719000Z",
  "data_removed": false,
  "error": null,
  "id": "q8ztkpc3gxrj20cnmbrtye6s9r",
  "input": {
    "image": "https://replicate.delivery/pbxt/MfpzR98ZJmwZM7oUcVqeFNW5DUvtTGFDTopDvFC40JdGhZKB/i2v_input.JPG",
    "prompt": "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline's intricate details and the refreshing atmosphere of the seaside.",
    "max_area": "1280*720",
    "num_frames": 81,
    "sample_shift": 5,
    "sample_steps": 30,
    "frames_per_second": 16,
    "sample_guide_scale": 5
  },
  "logs": "输入图像尺寸: 832x1104\n使用随机种子: 2082605660\n开始生成视频...\n参数: 提示词长度=496, 最大面积=1280*720, 帧数=81\n采样步数=30, 引导尺度=5.0, 偏移因子=5.0\n/src/wan/modules/clip.py:540: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.\nwith torch.cuda.amp.autocast(dtype=self.dtype):\n/src/wan/modules/vae.py:651: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.\nwith amp.autocast(dtype=self.dtype):\n/src/wan/image2video.py:255: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.\nwith amp.autocast(dtype=self.param_dtype), torch.no_grad(), no_sync():\n  0%|          | 0/30 [00:00<?, ?it/s]/src/wan/modules/model.py:535: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.\nwith amp.autocast(dtype=torch.float32):\n/src/wan/modules/model.py:293: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.\nwith amp.autocast(dtype=torch.float32):\n/src/wan/modules/model.py:301: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.\nwith amp.autocast(dtype=torch.float32):\n/src/wan/modules/model.py:308: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.\nwith amp.autocast(dtype=torch.float32):\n/src/wan/modules/model.py:340: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.\nwith amp.autocast(dtype=torch.float32):\n  3%|▎         | 1/30 [01:06<32:16, 66.77s/it]\n  7%|▋         | 2/30 [02:14<31:19, 67.11s/it]\n 10%|█         | 3/30 [03:21<30:15, 67.25s/it]\n 13%|█▎        | 4/30 [04:29<29:10, 67.34s/it]\n 17%|█▋        | 5/30 [05:36<28:03, 67.35s/it]\n 20%|██        | 6/30 [06:43<26:57, 67.38s/it]\n 23%|██▎       | 7/30 [07:51<25:49, 67.39s/it]\n 27%|██▋       | 8/30 [08:58<24:42, 67.38s/it]\n 30%|███       | 9/30 [10:06<23:35, 67.40s/it]\n 33%|███▎      | 10/30 [11:13<22:27, 67.40s/it]\n 37%|███▋      | 11/30 [12:20<21:20, 67.39s/it]\n 40%|████      | 12/30 [13:28<20:12, 67.37s/it]\n 43%|████▎     | 13/30 [14:35<19:05, 67.36s/it]\n 47%|████▋     | 14/30 [15:42<17:57, 67.34s/it]\n 50%|█████     | 15/30 [16:50<16:50, 67.35s/it]\n 53%|█████▎    | 16/30 [17:57<15:42, 67.34s/it]\n 57%|█████▋    | 17/30 [19:04<14:35, 67.33s/it]\n 60%|██████    | 18/30 [20:12<13:27, 67.33s/it]\n 63%|██████▎   | 19/30 [21:19<12:20, 67.32s/it]\n 67%|██████▋   | 20/30 [22:26<11:13, 67.31s/it]\n 70%|███████   | 21/30 [23:33<10:05, 67.31s/it]\n 73%|███████▎  | 22/30 [24:41<08:58, 67.32s/it]\n 77%|███████▋  | 23/30 [25:48<07:51, 67.32s/it]\n 80%|████████  | 24/30 [26:55<06:43, 67.33s/it]\n 83%|████████▎ | 25/30 [28:03<05:36, 67.31s/it]\n 87%|████████▋ | 26/30 [29:10<04:29, 67.30s/it]\n 90%|█████████ | 27/30 [30:17<03:21, 67.29s/it]\n 93%|█████████▎| 28/30 [31:25<02:14, 67.28s/it]\n 97%|█████████▋| 29/30 [32:32<01:07, 67.28s/it]\n100%|██████████| 30/30 [33:39<00:00, 67.27s/it]\n100%|██████████| 30/30 [33:39<00:00, 67.32s/it]\n/src/wan/modules/vae.py:658: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.\nwith amp.autocast(dtype=self.dtype):\n保存视频到: output_video.mp4, 帧率: 16 fps",
  "metrics": {
    "predict_time": 2098.768213578,
    "total_time": 2777.128319
  },
  "output": "https://replicate.delivery/yhqm/2eeOEKeH0XGXdph2HTtPNWRMoD0wxvjeD4TO1ydUDQuzfDLjC/output_video.mp4",
  "started_at": "2025-03-17T07:22:50.079105Z",
  "status": "succeeded",
  "urls": {
    "stream": "https://stream.replicate.com/v1/files/yswh-2z327spxaa2qkna2kux3fmuygsoh34xgliaz3ivwxz6dxgtpedqq",
    "get": "https://api.replicate.com/v1/predictions/q8ztkpc3gxrj20cnmbrtye6s9r",
    "cancel": "https://api.replicate.com/v1/predictions/q8ztkpc3gxrj20cnmbrtye6s9r/cancel"
  },
  "version": "d6a220d6990c8e166d4639a7b8fc9ef744e8e5297af2aef18edde1477140782e"
}

Generated in

34 minutes 59 seconds

Tweak it ShareReport

输入图像尺寸: 832x1104
使用随机种子: 2082605660
开始生成视频...
参数: 提示词长度=496, 最大面积=1280*720, 帧数=81
采样步数=30, 引导尺度=5.0, 偏移因子=5.0
/src/wan/modules/clip.py:540: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
with torch.cuda.amp.autocast(dtype=self.dtype):
/src/wan/modules/vae.py:651: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
with amp.autocast(dtype=self.dtype):
/src/wan/image2video.py:255: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
with amp.autocast(dtype=self.param_dtype), torch.no_grad(), no_sync():
  0%|          | 0/30 [00:00<?, ?it/s]/src/wan/modules/model.py:535: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
with amp.autocast(dtype=torch.float32):
/src/wan/modules/model.py:293: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
with amp.autocast(dtype=torch.float32):
/src/wan/modules/model.py:301: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
with amp.autocast(dtype=torch.float32):
/src/wan/modules/model.py:308: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
with amp.autocast(dtype=torch.float32):
/src/wan/modules/model.py:340: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
with amp.autocast(dtype=torch.float32):
  3%|▎         | 1/30 [01:06<32:16, 66.77s/it]
  7%|▋         | 2/30 [02:14<31:19, 67.11s/it]
 10%|█         | 3/30 [03:21<30:15, 67.25s/it]
 13%|█▎        | 4/30 [04:29<29:10, 67.34s/it]
 17%|█▋        | 5/30 [05:36<28:03, 67.35s/it]
 20%|██        | 6/30 [06:43<26:57, 67.38s/it]
 23%|██▎       | 7/30 [07:51<25:49, 67.39s/it]
 27%|██▋       | 8/30 [08:58<24:42, 67.38s/it]
 30%|███       | 9/30 [10:06<23:35, 67.40s/it]
 33%|███▎      | 10/30 [11:13<22:27, 67.40s/it]
 37%|███▋      | 11/30 [12:20<21:20, 67.39s/it]
 40%|████      | 12/30 [13:28<20:12, 67.37s/it]
 43%|████▎     | 13/30 [14:35<19:05, 67.36s/it]
 47%|████▋     | 14/30 [15:42<17:57, 67.34s/it]
 50%|█████     | 15/30 [16:50<16:50, 67.35s/it]
 53%|█████▎    | 16/30 [17:57<15:42, 67.34s/it]
 57%|█████▋    | 17/30 [19:04<14:35, 67.33s/it]
 60%|██████    | 18/30 [20:12<13:27, 67.33s/it]
 63%|██████▎   | 19/30 [21:19<12:20, 67.32s/it]
 67%|██████▋   | 20/30 [22:26<11:13, 67.31s/it]
 70%|███████   | 21/30 [23:33<10:05, 67.31s/it]
 73%|███████▎  | 22/30 [24:41<08:58, 67.32s/it]
 77%|███████▋  | 23/30 [25:48<07:51, 67.32s/it]
 80%|████████  | 24/30 [26:55<06:43, 67.33s/it]
 83%|████████▎ | 25/30 [28:03<05:36, 67.31s/it]
 87%|████████▋ | 26/30 [29:10<04:29, 67.30s/it]
 90%|█████████ | 27/30 [30:17<03:21, 67.29s/it]
 93%|█████████▎| 28/30 [31:25<02:14, 67.28s/it]
 97%|█████████▋| 29/30 [32:32<01:07, 67.28s/it]
100%|██████████| 30/30 [33:39<00:00, 67.27s/it]
100%|██████████| 30/30 [33:39<00:00, 67.32s/it]
/src/wan/modules/vae.py:658: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
with amp.autocast(dtype=self.dtype):
保存视频到: output_video.mp4, 帧率: 16 fps

Examples

View more examples

Run time and cost

This model runs on Nvidia A100 (80GB) GPU hardware. We don't yet have enough runs of this model to provide performance information.

Readme

This model doesn't have a readme.