jyoung105/instant-style | Run with an API on Replicate

jyoung105 / instant-style

Free Lunch towards Style-Preserving in Text-to-Image Generation by InstantX team (Updated 1 year, 2 months ago)

Cold

Public
1.9K runs
L40S
GitHub
Paper

Iterate in playground

Run with an API

Playground API Examples README Versions

Input

style_image

file

Input reference image for style

prompt

string

Shift + Return to add a new line

Input prompt

negative_prompt

string

Shift + Return to add a new line

Input negative prompt

width

integer

(minimum: 512, maximum: 2048)

Width of output image

Default: 1024

height

integer

(minimum: 512, maximum: 2048)

Height of output image

Default: 1024

num_outputs

integer

(minimum: 1, maximum: 4)

Number of output images

Default: 1

num_inference_steps

integer

(minimum: 1, maximum: 50)

Number of denoising steps

Default: 30

guidance_scale

number

(minimum: 1, maximum: 20)

Scale for classifier-free guidance

Default: 5

seed

integer

Random seed. Leave blank to randomize the seed

style_strength

number

(minimum: 0, maximum: 2)

Conditioning scale for ip-adapter

Default: 1

block_mode

string

Mode to reference the image: original, style with or without layout

Default: "style-only"

adapter_mode

string

Mode to reference the image: high flexibility but low fidelity or low flexibility but high fidelity

Default: "original"

enable_negative_content

boolean

Choose whether you extract only style part of prompt or not

Default: false

negative_content

string

Shift + Return to add a new line

Input negative prompt

negative_content_strength

number

(minimum: 0, maximum: 1)

Conditioning scale for content you want to exclude

Default: 0.5

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

npx create-replicate --model=jyoung105/instant-style

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 jyoung105/instant-style using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "jyoung105/instant-style:c6f01e12f31cb99f9ee774a78992a71294f630a6f433d9aecfdc33b816fc4baa",
  {
    input: {
      width: 1024,
      height: 1024,
      prompt: "A dog",
      block_mode: "style-only",
      num_outputs: 1,
      style_image: "https://replicate.delivery/pbxt/KgoIir7Dz5vLdvy1ZOspa1dVI8L6FldaTuGH4d2QYDx1NGjl/0.jpg",
      adapter_mode: "original",
      guidance_scale: 5,
      style_strength: 1,
      num_inference_steps: 30,
      enable_negative_content: false,
      negative_content_strength: 0.5
    }
  }
);

// To access the file URL:
console.log(output[0].url()); //=> "http://example.com"

// To write the file to disk:
fs.writeFile("my-image.png", output[0]);

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 jyoung105/instant-style using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "jyoung105/instant-style:c6f01e12f31cb99f9ee774a78992a71294f630a6f433d9aecfdc33b816fc4baa",
    input={
        "width": 1024,
        "height": 1024,
        "prompt": "A dog",
        "block_mode": "style-only",
        "num_outputs": 1,
        "style_image": "https://replicate.delivery/pbxt/KgoIir7Dz5vLdvy1ZOspa1dVI8L6FldaTuGH4d2QYDx1NGjl/0.jpg",
        "adapter_mode": "original",
        "guidance_scale": 5,
        "style_strength": 1,
        "num_inference_steps": 30,
        "enable_negative_content": False,
        "negative_content_strength": 0.5
    }
)
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 jyoung105/instant-style 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": "jyoung105/instant-style:c6f01e12f31cb99f9ee774a78992a71294f630a6f433d9aecfdc33b816fc4baa",
    "input": {
      "width": 1024,
      "height": 1024,
      "prompt": "A dog",
      "block_mode": "style-only",
      "num_outputs": 1,
      "style_image": "https://replicate.delivery/pbxt/KgoIir7Dz5vLdvy1ZOspa1dVI8L6FldaTuGH4d2QYDx1NGjl/0.jpg",
      "adapter_mode": "original",
      "guidance_scale": 5,
      "style_strength": 1,
      "num_inference_steps": 30,
      "enable_negative_content": false,
      "negative_content_strength": 0.5
    }
  }' \
  https://api.replicate.com/v1/predictions

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

Output

{
  "completed_at": "2024-04-07T18:32:47.332133Z",
  "created_at": "2024-04-07T18:29:48.706000Z",
  "data_removed": false,
  "error": null,
  "id": "sd42yq7049rgg0ceq79995gf3g",
  "input": {
    "width": 1024,
    "height": 1024,
    "prompt": "A dog",
    "block_mode": "style-only",
    "num_outputs": 1,
    "style_image": "https://replicate.delivery/pbxt/KgoIir7Dz5vLdvy1ZOspa1dVI8L6FldaTuGH4d2QYDx1NGjl/0.jpg",
    "adapter_mode": "original",
    "guidance_scale": 5,
    "style_strength": 1,
    "num_inference_steps": 30,
    "enable_negative_content": false,
    "negative_content_strength": 0.5
  },
  "logs": "Using seed: 21469\nDEVICE: cuda\nDTYPE: torch.float16\nStart inference...\n[Debug] Prompt: A dog, best quality, high detail, sharp focus,\n[Debug] Neg Prompt: text, watermark, lowres, low quality, worst quality, deformed, glitch, low contrast, noisy, saturation, blurry,\n  0%|          | 0/30 [00:00<?, ?it/s]\n  3%|▎         | 1/30 [00:00<00:07,  4.03it/s]\n  7%|▋         | 2/30 [00:00<00:04,  5.93it/s]\n 10%|█         | 3/30 [00:00<00:04,  5.51it/s]\n 13%|█▎        | 4/30 [00:00<00:04,  5.33it/s]\n 17%|█▋        | 5/30 [00:00<00:04,  5.22it/s]\n 20%|██        | 6/30 [00:01<00:04,  5.14it/s]\n 23%|██▎       | 7/30 [00:01<00:04,  5.11it/s]\n 27%|██▋       | 8/30 [00:01<00:04,  5.09it/s]\n 30%|███       | 9/30 [00:01<00:04,  5.08it/s]\n 33%|███▎      | 10/30 [00:01<00:03,  5.07it/s]\n 37%|███▋      | 11/30 [00:02<00:03,  5.06it/s]\n 40%|████      | 12/30 [00:02<00:03,  5.04it/s]\n 43%|████▎     | 13/30 [00:02<00:03,  5.03it/s]\n 47%|████▋     | 14/30 [00:02<00:03,  5.05it/s]\n 50%|█████     | 15/30 [00:02<00:02,  5.05it/s]\n 53%|█████▎    | 16/30 [00:03<00:02,  5.04it/s]\n 57%|█████▋    | 17/30 [00:03<00:02,  5.03it/s]\n 60%|██████    | 18/30 [00:03<00:02,  5.03it/s]\n 63%|██████▎   | 19/30 [00:03<00:02,  5.03it/s]\n 67%|██████▋   | 20/30 [00:03<00:01,  5.04it/s]\n 70%|███████   | 21/30 [00:04<00:01,  5.03it/s]\n 73%|███████▎  | 22/30 [00:04<00:01,  5.02it/s]\n 77%|███████▋  | 23/30 [00:04<00:01,  5.02it/s]\n 80%|████████  | 24/30 [00:04<00:01,  5.02it/s]\n 83%|████████▎ | 25/30 [00:04<00:00,  5.01it/s]\n 87%|████████▋ | 26/30 [00:05<00:00,  5.03it/s]\n 90%|█████████ | 27/30 [00:05<00:00,  5.02it/s]\n 93%|█████████▎| 28/30 [00:05<00:00,  5.02it/s]\n 97%|█████████▋| 29/30 [00:05<00:00,  5.02it/s]\n100%|██████████| 30/30 [00:05<00:00,  5.03it/s]\n100%|██████████| 30/30 [00:05<00:00,  5.06it/s]\nFinish generation in 6.64692759513855 secs.",
  "metrics": {
    "predict_time": 17.960438,
    "total_time": 178.626133
  },
  "output": [
    "https://replicate.delivery/pbxt/FzfZtZfRRUnQgU78LexGzwyH59WDlQ33zRQXUrgnQXgbGTQlA/out_0.png"
  ],
  "started_at": "2024-04-07T18:32:29.371695Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/sd42yq7049rgg0ceq79995gf3g",
    "cancel": "https://api.replicate.com/v1/predictions/sd42yq7049rgg0ceq79995gf3g/cancel"
  },
  "version": "a11eb1cb58caf320176953852871ac4882c49baeb88893cd1eacfcd91cd71769"
}

Generated in

18.0 seconds

Tweak it ShareReport View full prediction

Using seed: 21469
DEVICE: cuda
DTYPE: torch.float16
Start inference...
[Debug] Prompt: A dog, best quality, high detail, sharp focus,
[Debug] Neg Prompt: text, watermark, lowres, low quality, worst quality, deformed, glitch, low contrast, noisy, saturation, blurry,
  0%|          | 0/30 [00:00<?, ?it/s]
  3%|▎         | 1/30 [00:00<00:07,  4.03it/s]
  7%|▋         | 2/30 [00:00<00:04,  5.93it/s]
 10%|█         | 3/30 [00:00<00:04,  5.51it/s]
 13%|█▎        | 4/30 [00:00<00:04,  5.33it/s]
 17%|█▋        | 5/30 [00:00<00:04,  5.22it/s]
 20%|██        | 6/30 [00:01<00:04,  5.14it/s]
 23%|██▎       | 7/30 [00:01<00:04,  5.11it/s]
 27%|██▋       | 8/30 [00:01<00:04,  5.09it/s]
 30%|███       | 9/30 [00:01<00:04,  5.08it/s]
 33%|███▎      | 10/30 [00:01<00:03,  5.07it/s]
 37%|███▋      | 11/30 [00:02<00:03,  5.06it/s]
 40%|████      | 12/30 [00:02<00:03,  5.04it/s]
 43%|████▎     | 13/30 [00:02<00:03,  5.03it/s]
 47%|████▋     | 14/30 [00:02<00:03,  5.05it/s]
 50%|█████     | 15/30 [00:02<00:02,  5.05it/s]
 53%|█████▎    | 16/30 [00:03<00:02,  5.04it/s]
 57%|█████▋    | 17/30 [00:03<00:02,  5.03it/s]
 60%|██████    | 18/30 [00:03<00:02,  5.03it/s]
 63%|██████▎   | 19/30 [00:03<00:02,  5.03it/s]
 67%|██████▋   | 20/30 [00:03<00:01,  5.04it/s]
 70%|███████   | 21/30 [00:04<00:01,  5.03it/s]
 73%|███████▎  | 22/30 [00:04<00:01,  5.02it/s]
 77%|███████▋  | 23/30 [00:04<00:01,  5.02it/s]
 80%|████████  | 24/30 [00:04<00:01,  5.02it/s]
 83%|████████▎ | 25/30 [00:04<00:00,  5.01it/s]
 87%|████████▋ | 26/30 [00:05<00:00,  5.03it/s]
 90%|█████████ | 27/30 [00:05<00:00,  5.02it/s]
 93%|█████████▎| 28/30 [00:05<00:00,  5.02it/s]
 97%|█████████▋| 29/30 [00:05<00:00,  5.02it/s]
100%|██████████| 30/30 [00:05<00:00,  5.03it/s]
100%|██████████| 30/30 [00:05<00:00,  5.06it/s]
Finish generation in 6.64692759513855 secs.

This output was created using a different version of the model, jyoung105/instant-style:a11eb1cb.

Examples

View more examples

Run time and cost

This model costs approximately $0.12 to run on Replicate, or 8 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 128 seconds. The predict time for this model varies significantly based on the inputs.

Readme

This model doesn't have a readme.