Input

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

npx create-replicate --model=tuannha/instant-character

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 tuannha/instant-character using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "tuannha/instant-character:df5eed34fa9c812acf62d3ca79874daf9b5e78c2bee11f4ada182a55dd5c1712",
  {
    input: {
      lora: "ghibli_style",
      seed: -1,
      width: 768,
      height: 1344,
      prompt: "a character in the library",
      subject_image: "https://replicate.delivery/pbxt/Mr9015ajTj5WrUEuW6YyWbxxRifCNCo0SbiA0L0F4vqjQmXS/face.png",
      guidance_scale: 3.5,
      num_inference_steps: 28
    }
  }
);

// 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 tuannha/instant-character using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "tuannha/instant-character:df5eed34fa9c812acf62d3ca79874daf9b5e78c2bee11f4ada182a55dd5c1712",
    input={
        "lora": "ghibli_style",
        "seed": -1,
        "width": 768,
        "height": 1344,
        "prompt": "a character in the library",
        "subject_image": "https://replicate.delivery/pbxt/Mr9015ajTj5WrUEuW6YyWbxxRifCNCo0SbiA0L0F4vqjQmXS/face.png",
        "guidance_scale": 3.5,
        "num_inference_steps": 28
    }
)
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 tuannha/instant-character 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": "tuannha/instant-character:df5eed34fa9c812acf62d3ca79874daf9b5e78c2bee11f4ada182a55dd5c1712",
    "input": {
      "lora": "ghibli_style",
      "seed": -1,
      "width": 768,
      "height": 1344,
      "prompt": "a character in the library",
      "subject_image": "https://replicate.delivery/pbxt/Mr9015ajTj5WrUEuW6YyWbxxRifCNCo0SbiA0L0F4vqjQmXS/face.png",
      "guidance_scale": 3.5,
      "num_inference_steps": 28
    }
  }' \
  https://api.replicate.com/v1/predictions

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

Output

{
  "completed_at": "2025-04-18T03:40:52.931853Z",
  "created_at": "2025-04-18T03:40:33.078000Z",
  "data_removed": false,
  "error": null,
  "id": "2vxx06evpsrm80cp8vxsffjj5c",
  "input": {
    "lora": "ghibli_style",
    "seed": -1,
    "width": 768,
    "height": 1344,
    "prompt": "a character in the library",
    "subject_image": "https://replicate.delivery/pbxt/Mr9015ajTj5WrUEuW6YyWbxxRifCNCo0SbiA0L0F4vqjQmXS/face.png",
    "guidance_scale": 3.5,
    "num_inference_steps": 28
  },
  "logs": "loading lora in transformer ...:   0%|          | 0/988 [00:00<?, ?it/s]\nloading lora in transformer ...:  53%|█████▎    | 528/988 [00:00<00:00, 5258.64it/s]\nloading lora in transformer ...: 100%|██████████| 988/988 [00:00<00:00, 3879.36it/s]\nloading lora in text_encoder ...: 0it [00:00, ?it/s]\nloading lora in text_encoder ...: 0it [00:00, ?it/s]\n  0%|          | 0/28 [00:00<?, ?it/s]\n  4%|▎         | 1/28 [00:00<00:17,  1.58it/s]\n  7%|▋         | 2/28 [00:01<00:13,  1.87it/s]\n 11%|█         | 3/28 [00:01<00:14,  1.73it/s]\n 14%|█▍        | 4/28 [00:02<00:14,  1.67it/s]\n 18%|█▊        | 5/28 [00:02<00:14,  1.63it/s]\n 21%|██▏       | 6/28 [00:03<00:13,  1.61it/s]\n 25%|██▌       | 7/28 [00:04<00:13,  1.60it/s]\n 29%|██▊       | 8/28 [00:04<00:12,  1.59it/s]\n 32%|███▏      | 9/28 [00:05<00:11,  1.59it/s]\n 36%|███▌      | 10/28 [00:06<00:11,  1.58it/s]\n 39%|███▉      | 11/28 [00:06<00:10,  1.58it/s]\n 43%|████▎     | 12/28 [00:07<00:10,  1.58it/s]\n 46%|████▋     | 13/28 [00:08<00:09,  1.58it/s]\n 50%|█████     | 14/28 [00:08<00:08,  1.58it/s]\n 54%|█████▎    | 15/28 [00:09<00:08,  1.57it/s]\n 57%|█████▋    | 16/28 [00:09<00:07,  1.57it/s]\n 61%|██████    | 17/28 [00:10<00:06,  1.57it/s]\n 64%|██████▍   | 18/28 [00:11<00:06,  1.57it/s]\n 68%|██████▊   | 19/28 [00:11<00:05,  1.57it/s]\n 71%|███████▏  | 20/28 [00:12<00:05,  1.57it/s]\n 75%|███████▌  | 21/28 [00:13<00:04,  1.57it/s]\n 79%|███████▊  | 22/28 [00:13<00:03,  1.57it/s]\n 82%|████████▏ | 23/28 [00:14<00:03,  1.57it/s]\n 86%|████████▌ | 24/28 [00:15<00:02,  1.56it/s]\n 89%|████████▉ | 25/28 [00:15<00:01,  1.56it/s]\n 93%|█████████▎| 26/28 [00:16<00:01,  1.56it/s]\n 96%|█████████▋| 27/28 [00:17<00:00,  1.56it/s]\n100%|██████████| 28/28 [00:17<00:00,  1.56it/s]\n100%|██████████| 28/28 [00:17<00:00,  1.59it/s]\nloading lora in transformer ...:   0%|          | 0/988 [00:00<?, ?it/s]\nloading lora in transformer ...:  65%|██████▍   | 640/988 [00:00<00:00, 6389.53it/s]\nloading lora in transformer ...: 100%|██████████| 988/988 [00:00<00:00, 4527.79it/s]\nloading lora in text_encoder ...: 0it [00:00, ?it/s]\nloading lora in text_encoder ...: 0it [00:00, ?it/s]",
  "metrics": {
    "predict_time": 19.846660952,
    "total_time": 19.853853
  },
  "output": "https://replicate.delivery/xezq/wtf5hwaR2EwgVaEsMDU41pBCcjntw5wVOG99ujj9NPWi37RKA/73dbbbd9-314a-4cd8-8022-cb77a6ae877b.png",
  "started_at": "2025-04-18T03:40:33.085192Z",
  "status": "succeeded",
  "urls": {
    "stream": "https://stream.replicate.com/v1/files/bcwr-np7dnpvqloa3gsnjhwf3yxlha7kmsdujlxxhg4li7gixiobwpkva",
    "get": "https://api.replicate.com/v1/predictions/2vxx06evpsrm80cp8vxsffjj5c",
    "cancel": "https://api.replicate.com/v1/predictions/2vxx06evpsrm80cp8vxsffjj5c/cancel"
  },
  "version": "cf2799a4d552fd49b3ef7d58cbcbb147d6f098ec42a307e2e857955e35c7bd6f"
}

Generated in

19.9 seconds

Tweak it Share Report

loading lora in transformer ...:   0%|          | 0/988 [00:00<?, ?it/s]
loading lora in transformer ...:  53%|█████▎    | 528/988 [00:00<00:00, 5258.64it/s]
loading lora in transformer ...: 100%|██████████| 988/988 [00:00<00:00, 3879.36it/s]
loading lora in text_encoder ...: 0it [00:00, ?it/s]
loading lora in text_encoder ...: 0it [00:00, ?it/s]
  0%|          | 0/28 [00:00<?, ?it/s]
  4%|▎         | 1/28 [00:00<00:17,  1.58it/s]
  7%|▋         | 2/28 [00:01<00:13,  1.87it/s]
 11%|█         | 3/28 [00:01<00:14,  1.73it/s]
 14%|█▍        | 4/28 [00:02<00:14,  1.67it/s]
 18%|█▊        | 5/28 [00:02<00:14,  1.63it/s]
 21%|██▏       | 6/28 [00:03<00:13,  1.61it/s]
 25%|██▌       | 7/28 [00:04<00:13,  1.60it/s]
 29%|██▊       | 8/28 [00:04<00:12,  1.59it/s]
 32%|███▏      | 9/28 [00:05<00:11,  1.59it/s]
 36%|███▌      | 10/28 [00:06<00:11,  1.58it/s]
 39%|███▉      | 11/28 [00:06<00:10,  1.58it/s]
 43%|████▎     | 12/28 [00:07<00:10,  1.58it/s]
 46%|████▋     | 13/28 [00:08<00:09,  1.58it/s]
 50%|█████     | 14/28 [00:08<00:08,  1.58it/s]
 54%|█████▎    | 15/28 [00:09<00:08,  1.57it/s]
 57%|█████▋    | 16/28 [00:09<00:07,  1.57it/s]
 61%|██████    | 17/28 [00:10<00:06,  1.57it/s]
 64%|██████▍   | 18/28 [00:11<00:06,  1.57it/s]
 68%|██████▊   | 19/28 [00:11<00:05,  1.57it/s]
 71%|███████▏  | 20/28 [00:12<00:05,  1.57it/s]
 75%|███████▌  | 21/28 [00:13<00:04,  1.57it/s]
 79%|███████▊  | 22/28 [00:13<00:03,  1.57it/s]
 82%|████████▏ | 23/28 [00:14<00:03,  1.57it/s]
 86%|████████▌ | 24/28 [00:15<00:02,  1.56it/s]
 89%|████████▉ | 25/28 [00:15<00:01,  1.56it/s]
 93%|█████████▎| 26/28 [00:16<00:01,  1.56it/s]
 96%|█████████▋| 27/28 [00:17<00:00,  1.56it/s]
100%|██████████| 28/28 [00:17<00:00,  1.56it/s]
100%|██████████| 28/28 [00:17<00:00,  1.59it/s]
loading lora in transformer ...:   0%|          | 0/988 [00:00<?, ?it/s]
loading lora in transformer ...:  65%|██████▍   | 640/988 [00:00<00:00, 6389.53it/s]
loading lora in transformer ...: 100%|██████████| 988/988 [00:00<00:00, 4527.79it/s]
loading lora in text_encoder ...: 0it [00:00, ?it/s]
loading lora in text_encoder ...: 0it [00:00, ?it/s]

This output was created using a different version of the model, tuannha/instant-character:cf2799a4.

Examples

View more examples

Run time and cost

This model costs approximately $0.089 to run on Replicate, or 11 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 92 seconds. The predict time for this model varies significantly based on the inputs.

Readme

InstantCharacter: Personalize Any Characters with a Scalable Diffusion Transformer Framework

InstantCharacter is an innovative, tuning-free method designed to achieve character-preserving generation from a single image, supporting a variety of downstream tasks.

Release

[2025/04/18] 🔥 We release the demo checkpoints and code.

[2025/04/02] 🔥 We launch the project page.

Download

You can directly download the model from Huggingface.

huggingface-cli download --resume-download Tencent/InstantCharacter --local-dir checkpoints --local-dir-use-symlinks False

If you cannot access to Huggingface, you can use hf-mirror to download models.

export HF_ENDPOINT=https://hf-mirror.com
huggingface-cli download --resume-download Tencent/InstantCharacter --local-dir checkpoints --local-dir-use-symlinks False

tuannha / instant-character

Input

Output

Examples

Run time and cost

Readme

InstantCharacter: Personalize Any Characters with a Scalable Diffusion Transformer Framework

Release

Download

Logs (2vxx06evpsrm80cp8vxsffjj5c)