fofr/kolors – Run with an API on Replicate

fofr / kolors

A large-scale text-to-image generation model based on latent diffusion, developed by the Kuaishou Kolors team

Cold

Public
30.3K runs
L40S
GitHub
Paper
License

Run with an API

Playground API Examples README Versions

Input

prompt

string

Shift + Return to add a new line

a kingfisher saying "hello" in a speech bubblea kingfisher saying "hello" in a speech bubble

Default: ""

negative_prompt

string

Shift + Return to add a new line

Things you do not want to see in your image

Default: ""

number_of_images

integer

(minimum: 1, maximum: 10)

Number of images to generate

Default: 1

width

integer

(minimum: 512, maximum: 2048)

Width of the image

Default: 1024

height

integer

(minimum: 512, maximum: 2048)

Height of the image

Default: 1024

steps

integer

(minimum: 1, maximum: 50)

Number of inference steps

Default: 25

cfg

number

(minimum: 0, maximum: 20)

Guidance scale

Default: 5

scheduler

string

Scheduler

Default: "EulerDiscreteScheduler"

output_format

string

Format of the output images

Default: "webp"

output_quality

integer

(minimum: 0, maximum: 100)

Quality of the output images, from 0 to 100. 100 is best quality, 0 is lowest quality.

Default: 80

seed

integer

Set a seed for reproducibility. Random by default.

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

npx create-replicate --model=fofr/kolors

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 fofr/kolors using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "fofr/kolors:6145c9c5cbd646873232d661fe0398ab87ac93db66d5c365959a471c651862d7",
  {
    input: {
      cfg: 5,
      steps: 25,
      width: 1024,
      height: 1024,
      prompt: "a kingfisher saying \"hello\" in a speech bubble",
      scheduler: "EulerDiscreteScheduler",
      output_format: "webp",
      output_quality: 80,
      negative_prompt: "",
      number_of_images: 1
    }
  }
);

// 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 fofr/kolors using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "fofr/kolors:6145c9c5cbd646873232d661fe0398ab87ac93db66d5c365959a471c651862d7",
    input={
        "cfg": 5,
        "steps": 25,
        "width": 1024,
        "height": 1024,
        "prompt": "a kingfisher saying \"hello\" in a speech bubble",
        "scheduler": "EulerDiscreteScheduler",
        "output_format": "webp",
        "output_quality": 80,
        "negative_prompt": "",
        "number_of_images": 1
    }
)
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 fofr/kolors 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": "6145c9c5cbd646873232d661fe0398ab87ac93db66d5c365959a471c651862d7",
    "input": {
      "cfg": 5,
      "steps": 25,
      "width": 1024,
      "height": 1024,
      "prompt": "a kingfisher saying \\"hello\\" in a speech bubble",
      "scheduler": "EulerDiscreteScheduler",
      "output_format": "webp",
      "output_quality": 80,
      "negative_prompt": "",
      "number_of_images": 1
    }
  }' \
  https://api.replicate.com/v1/predictions

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

Output

{
  "completed_at": "2024-07-08T15:07:09.536014Z",
  "created_at": "2024-07-08T15:07:03.668000Z",
  "data_removed": false,
  "error": null,
  "id": "cdz12h2gehrgg0cgjbh8kteym4",
  "input": {
    "cfg": 5,
    "steps": 25,
    "width": 1024,
    "height": 1024,
    "prompt": "a kingfisher saying \"hello\" in a speech bubble",
    "scheduler": "EulerDiscreteScheduler",
    "output_format": "webp",
    "output_quality": 80,
    "negative_prompt": "",
    "number_of_images": 1
  },
  "logs": "Random seed set to: 2209112854\nRunning workflow\ngot prompt\nExecuting node 2, title: Kolors Sampler, class type: KolorsSampler\n  0%|          | 0/25 [00:00<?, ?it/s]\n  8%|▊         | 2/25 [00:00<00:03,  7.09it/s]\n 12%|█▏        | 3/25 [00:00<00:03,  6.17it/s]\n 16%|█▌        | 4/25 [00:00<00:03,  5.76it/s]\n 20%|██        | 5/25 [00:00<00:03,  5.55it/s]\n 24%|██▍       | 6/25 [00:01<00:03,  5.40it/s]\n 28%|██▊       | 7/25 [00:01<00:03,  5.32it/s]\n 32%|███▏      | 8/25 [00:01<00:03,  5.27it/s]\n 36%|███▌      | 9/25 [00:01<00:03,  5.24it/s]\n 40%|████      | 10/25 [00:01<00:02,  5.22it/s]\n 44%|████▍     | 11/25 [00:02<00:02,  5.19it/s]\n 48%|████▊     | 12/25 [00:02<00:02,  5.18it/s]\n 52%|█████▏    | 13/25 [00:02<00:02,  5.17it/s]\n 56%|█████▌    | 14/25 [00:02<00:02,  5.17it/s]\n 60%|██████    | 15/25 [00:02<00:01,  5.17it/s]\n 64%|██████▍   | 16/25 [00:02<00:01,  5.16it/s]\n 68%|██████▊   | 17/25 [00:03<00:01,  5.15it/s]\n 72%|███████▏  | 18/25 [00:03<00:01,  5.15it/s]\n 76%|███████▌  | 19/25 [00:03<00:01,  5.15it/s]\n 80%|████████  | 20/25 [00:03<00:00,  5.15it/s]\n 84%|████████▍ | 21/25 [00:03<00:00,  5.15it/s]\n 88%|████████▊ | 22/25 [00:04<00:00,  5.15it/s]\n 92%|█████████▏| 23/25 [00:04<00:00,  5.15it/s]\n 96%|█████████▌| 24/25 [00:04<00:00,  5.15it/s]\n100%|██████████| 25/25 [00:04<00:00,  5.15it/s]\n100%|██████████| 25/25 [00:04<00:00,  5.27it/s]\nExecuting node 7, title: VAE Decode, class type: VAEDecode\nExecuting node 9, title: Save Image, class type: SaveImage\nPrompt executed in 5.38 seconds\noutputs:  {'9': {'images': [{'filename': 'R8_Kolor_00001_.png', 'subfolder': '', 'type': 'output'}]}}\n====================================\nR8_Kolor_00001_.png",
  "metrics": {
    "predict_time": 5.858137734,
    "total_time": 5.868014
  },
  "output": [
    "https://replicate.delivery/pbxt/D1xnP9anGMLMKdNgEz8IM8VRMxb1fDmm6waHBZ2YeAhdKbGTA/R8_Kolor_00001_.webp"
  ],
  "started_at": "2024-07-08T15:07:03.677876Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/cdz12h2gehrgg0cgjbh8kteym4",
    "cancel": "https://api.replicate.com/v1/predictions/cdz12h2gehrgg0cgjbh8kteym4/cancel"
  },
  "version": "05df67f7cf9d1cc4a07af44340d92aa48df89601a30e9b76523addcad3a00bca"
}

Generated in

5.9 seconds

Tweak it Share Report

Random seed set to: 2209112854
Running workflow
got prompt
Executing node 2, title: Kolors Sampler, class type: KolorsSampler
  0%|          | 0/25 [00:00<?, ?it/s]
  8%|▊         | 2/25 [00:00<00:03,  7.09it/s]
 12%|█▏        | 3/25 [00:00<00:03,  6.17it/s]
 16%|█▌        | 4/25 [00:00<00:03,  5.76it/s]
 20%|██        | 5/25 [00:00<00:03,  5.55it/s]
 24%|██▍       | 6/25 [00:01<00:03,  5.40it/s]
 28%|██▊       | 7/25 [00:01<00:03,  5.32it/s]
 32%|███▏      | 8/25 [00:01<00:03,  5.27it/s]
 36%|███▌      | 9/25 [00:01<00:03,  5.24it/s]
 40%|████      | 10/25 [00:01<00:02,  5.22it/s]
 44%|████▍     | 11/25 [00:02<00:02,  5.19it/s]
 48%|████▊     | 12/25 [00:02<00:02,  5.18it/s]
 52%|█████▏    | 13/25 [00:02<00:02,  5.17it/s]
 56%|█████▌    | 14/25 [00:02<00:02,  5.17it/s]
 60%|██████    | 15/25 [00:02<00:01,  5.17it/s]
 64%|██████▍   | 16/25 [00:02<00:01,  5.16it/s]
 68%|██████▊   | 17/25 [00:03<00:01,  5.15it/s]
 72%|███████▏  | 18/25 [00:03<00:01,  5.15it/s]
 76%|███████▌  | 19/25 [00:03<00:01,  5.15it/s]
 80%|████████  | 20/25 [00:03<00:00,  5.15it/s]
 84%|████████▍ | 21/25 [00:03<00:00,  5.15it/s]
 88%|████████▊ | 22/25 [00:04<00:00,  5.15it/s]
 92%|█████████▏| 23/25 [00:04<00:00,  5.15it/s]
 96%|█████████▌| 24/25 [00:04<00:00,  5.15it/s]
100%|██████████| 25/25 [00:04<00:00,  5.15it/s]
100%|██████████| 25/25 [00:04<00:00,  5.27it/s]
Executing node 7, title: VAE Decode, class type: VAEDecode
Executing node 9, title: Save Image, class type: SaveImage
Prompt executed in 5.38 seconds
outputs:  {'9': {'images': [{'filename': 'R8_Kolor_00001_.png', 'subfolder': '', 'type': 'output'}]}}
====================================
R8_Kolor_00001_.png

This output was created using a different version of the model, fofr/kolors:05df67f7.

Examples

View more examples

Run time and cost

This model costs approximately $0.044 to run on Replicate, or 22 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 46 seconds. The predict time for this model varies significantly based on the inputs.

Readme

Kolors is a large-scale text-to-image generation model based on latent diffusion, developed by the Kuaishou Kolors team. Trained on billions of text-image pairs, Kolors exhibits significant advantages over both open-source and proprietary models in visual quality, complex semantic accuracy, and text rendering for both Chinese and English characters. Furthermore, Kolors supports both Chinese and English inputs, demonstrating strong performance in understanding and generating Chinese-specific content. For more details, please refer to this technical report.

Paper: https://github.com/Kwai-Kolors/Kolors/blob/master/imgs/Kolors_paper.pdf
Weights: https://huggingface.co/Kwai-Kolors/Kolors
Weights license: https://huggingface.co/Kwai-Kolors/Kolors/blob/main/MODEL_LICENSE
ComfyUI custom node: https://github.com/kijai/ComfyUI-KwaiKolorsWrapper