adalab-ai/kandinsky_v2_2 | Run with an API on Replicate

adalab-ai / kandinsky_v2_2

Generate images with Kandinsky 2.2 - Mix text and images

Cold

Public
42.7K runs
A100 (80GB)
GitHub
License

Iterate in playground

Run with an API

Playground API Examples README Versions

Input

task

string

Choose a task

Default: "text2img"

scheduler

string

Choose a scheduler

Default: "unipc"

prompt

string

Shift + Return to add a new line

A alien cheeseburger creature eating itself, claymation, cinematic, moody lightingA alien cheeseburger creature eating itself, claymation, cinematic, moody lighting

Provide input prompt

Default: "A alien cheeseburger creature eating itself, claymation, cinematic, moody lighting"

negative_prompt

string

Shift + Return to add a new line

ugly, tiling, oversaturated, undersaturated, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, extra limbs, disfigured, deformed, body out of frame, blurry, bad anatomy, blurred, watermark, grainy, signature, cut off, draftugly, tiling, oversaturated, undersaturated, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, extra limbs, disfigured, deformed, body out of frame, blurry, bad anatomy, blurred, watermark, grainy, signature, cut off, draft

Specify things to not see in the output for text2img and text_guided_img2img tasks

Default: "ugly, tiling, oversaturated, undersaturated, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, extra limbs, disfigured, deformed, body out of frame, blurry, bad anatomy, blurred, watermark, grainy, signature, cut off, draft"

image

file

Input image for text2img

width

integer

(minimum: 64, maximum: 1024)

Width of output image. Reduce the seeting if hits memory limits

Default: 512

height

integer

(minimum: 64, maximum: 1024)

Height of output image. Reduce the seeting if hits memory limits

Default: 512

num_outputs

integer

(minimum: 1, maximum: 4)

Number of images to output.

Default: 1

num_steps_prior

integer

(minimum: 1, maximum: 500)

Number of denoising steps in prior

Default: 2

num_inference_steps

integer

(minimum: 1, maximum: 500)

Number of denoising steps

Default: 18

guidance_scale

number

(minimum: 1, maximum: 10)

Scale for classifier-free guidance

Default: 4

seed

integer

Random seed. Leave blank to randomize the seed

img_weight

number

(minimum: 0, maximum: 10)

Weight of image - larger than 1 means more weight to image, lower than 0 is more weight to text

Default: 1

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

npx create-replicate --model=adalab-ai/kandinsky_v2_2

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 adalab-ai/kandinsky_v2_2 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "adalab-ai/kandinsky_v2_2:eb68c00d7fc7d7297a04dcecbdd29032361ab8f1d3f6843c32fbd6ec70532319",
  {
    input: {
      task: "text2img",
      width: 1024,
      height: 1024,
      prompt: "A alien cheeseburger creature eating itself, claymation, cinematic, moody lighting",
      scheduler: "unipc",
      img_weight: 1,
      num_outputs: 1,
      guidance_scale: 4,
      negative_prompt: "ugly, tiling, oversaturated, undersaturated, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, extra limbs, disfigured, deformed, body out of frame, blurry, bad anatomy, blurred, watermark, grainy, signature, cut off, draft",
      num_steps_prior: 2,
      num_inference_steps: 18
    }
  }
);

// 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 adalab-ai/kandinsky_v2_2 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "adalab-ai/kandinsky_v2_2:eb68c00d7fc7d7297a04dcecbdd29032361ab8f1d3f6843c32fbd6ec70532319",
    input={
        "task": "text2img",
        "width": 1024,
        "height": 1024,
        "prompt": "A alien cheeseburger creature eating itself, claymation, cinematic, moody lighting",
        "scheduler": "unipc",
        "img_weight": 1,
        "num_outputs": 1,
        "guidance_scale": 4,
        "negative_prompt": "ugly, tiling, oversaturated, undersaturated, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, extra limbs, disfigured, deformed, body out of frame, blurry, bad anatomy, blurred, watermark, grainy, signature, cut off, draft",
        "num_steps_prior": 2,
        "num_inference_steps": 18
    }
)
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 adalab-ai/kandinsky_v2_2 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": "adalab-ai/kandinsky_v2_2:eb68c00d7fc7d7297a04dcecbdd29032361ab8f1d3f6843c32fbd6ec70532319",
    "input": {
      "task": "text2img",
      "width": 1024,
      "height": 1024,
      "prompt": "A alien cheeseburger creature eating itself, claymation, cinematic, moody lighting",
      "scheduler": "unipc",
      "img_weight": 1,
      "num_outputs": 1,
      "guidance_scale": 4,
      "negative_prompt": "ugly, tiling, oversaturated, undersaturated, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, extra limbs, disfigured, deformed, body out of frame, blurry, bad anatomy, blurred, watermark, grainy, signature, cut off, draft",
      "num_steps_prior": 2,
      "num_inference_steps": 18
    }
  }' \
  https://api.replicate.com/v1/predictions

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

Output

{
  "completed_at": "2023-07-18T15:21:46.400543Z",
  "created_at": "2023-07-18T15:21:43.191497Z",
  "data_removed": false,
  "error": null,
  "id": "i5urxqbb5h7xc6p2tjwxyhbsw4",
  "input": {
    "task": "text2img",
    "width": 1024,
    "height": 1024,
    "prompt": "A alien cheeseburger creature eating itself, claymation, cinematic, moody lighting",
    "scheduler": "unipc",
    "img_weight": 1,
    "num_outputs": 1,
    "guidance_scale": 4,
    "negative_prompt": "ugly, tiling, oversaturated, undersaturated, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, extra limbs, disfigured, deformed, body out of frame, blurry, bad anatomy, blurred, watermark, grainy, signature, cut off, draft",
    "num_steps_prior": 2,
    "num_inference_steps": 18
  },
  "logs": "Using seed: 47117\n  0%|          | 0/2 [00:00<?, ?it/s]\n100%|██████████| 2/2 [00:00<00:00, 44.94it/s]\n  0%|          | 0/2 [00:00<?, ?it/s]\n100%|██████████| 2/2 [00:00<00:00, 45.88it/s]\n  0%|          | 0/18 [00:00<?, ?it/s]\n  6%|▌         | 1/18 [00:00<00:02,  6.39it/s]\n 17%|█▋        | 3/18 [00:00<00:01,  9.28it/s]\n 28%|██▊       | 5/18 [00:00<00:01, 10.58it/s]\n 39%|███▉      | 7/18 [00:00<00:00, 11.20it/s]\n 50%|█████     | 9/18 [00:00<00:00, 11.55it/s]\n 61%|██████    | 11/18 [00:00<00:00, 11.76it/s]\n 72%|███████▏  | 13/18 [00:01<00:00, 11.89it/s]\n 83%|████████▎ | 15/18 [00:01<00:00, 11.96it/s]\n 94%|█████████▍| 17/18 [00:01<00:00, 12.01it/s]\n100%|██████████| 18/18 [00:01<00:00, 11.45it/s]",
  "metrics": {
    "predict_time": 3.250161,
    "total_time": 3.209046
  },
  "output": [
    "https://replicate.delivery/pbxt/UWJyuHmKzaoRExhMNj0AUIxYkMGqjgr1zNOs6uWd6MfEAjoIA/out-0.png"
  ],
  "started_at": "2023-07-18T15:21:43.150382Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/i5urxqbb5h7xc6p2tjwxyhbsw4",
    "cancel": "https://api.replicate.com/v1/predictions/i5urxqbb5h7xc6p2tjwxyhbsw4/cancel"
  },
  "version": "eb68c00d7fc7d7297a04dcecbdd29032361ab8f1d3f6843c32fbd6ec70532319"
}

Generated in

3.3 seconds

Tweak itReport View full prediction

Examples

View more examples

Run time and cost

This model costs approximately $0.0019 to run on Replicate, or 526 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 A100 (80GB) GPU hardware. Predictions typically complete within 2 seconds. The predict time for this model varies significantly based on the inputs.

Readme

This model doesn't have a readme.