akisakurai/stable-karlo | Run with an API on Replicate

akisakurai / stable-karlo

Stable Diffusion Meets Karlo: a combination of the Karlo CLIP image embedding prior, and Stable Diffusion v2.1-768. (Updated 2 years, 2 months ago)

Cold

Public
996 runs
T4
GitHub

Iterate in playground

Run with an API

Playground API Examples README Versions

Input

prompt

string

Shift + Return to add a new line

fancy portraits of panda holding vast in laboratory with neon light and blue flame  by greg rutkowski, high detail, realistic, concept art, trending on artstation. fancy portraits of panda holding vast in laboratory with neon light and blue flame  by greg rutkowski, high detail, realistic, concept art, trending on artstation. 

Input prompt

Default: "A portrait of a panda holding a vial in a laboratory is trending on ArtStation. It was rendered in Blender and has an intriguing, cinematic quality with vivid colors, neon lights, and a blue flame. It evokes the feeling of a movie trailer."

negative_prompt

string

Shift + Return to add a new line

Specify things to not see in the output

Default: "amateur, mediocre, grainy, blur"

width

integer

Width of output image. Maximum size is 1024x768 or 768x1024 because of memory limits

Default: 768

height

integer

Height of output image. Maximum size is 1024x768 or 768x1024 because of memory limits

Default: 768

num_outputs

integer

(minimum: 1, maximum: 4)

Number of images to output.

Default: 1

num_inference_steps

integer

(minimum: 1, maximum: 500)

Number of denoising steps

Default: 50

prior_num_inference_steps

integer

(minimum: 1, maximum: 500)

Number of prior denoising steps

Default: 25

guidance_scale

number

(minimum: 1, maximum: 20)

Scale for classifier-free guidance

Default: 7.5

prior_guidance_scale

number

(minimum: 1, maximum: 20)

Scale for prior classifier-free guidance

Default: 4

noise_level

integer

The amount of noise to add to the image embeddings

Default: 0

scheduler

string

Choose a scheduler.

Default: "K_EULER_ANCESTRAL"

seed

integer

Random seed. Leave blank to randomize the seed

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

npx create-replicate --model=akisakurai/stable-karlo

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 akisakurai/stable-karlo using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "akisakurai/stable-karlo:92a88275607a45c6b38503de522937131a43003586f5b0cb227e28f64d139af7",
  {
    input: {
      width: 768,
      height: 768,
      prompt: "fancy portraits of panda holding vast in laboratory with neon light and blue flame  by greg rutkowski, high detail, realistic, concept art, trending on artstation. ",
      scheduler: "K_EULER_ANCESTRAL",
      noise_level: 0,
      num_outputs: 1,
      guidance_scale: 7.5,
      negative_prompt: "amateur, mediocre, grainy, blur",
      num_inference_steps: 50,
      prior_guidance_scale: 4,
      prior_num_inference_steps: 25
    }
  }
);

// 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 akisakurai/stable-karlo using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "akisakurai/stable-karlo:92a88275607a45c6b38503de522937131a43003586f5b0cb227e28f64d139af7",
    input={
        "width": 768,
        "height": 768,
        "prompt": "fancy portraits of panda holding vast in laboratory with neon light and blue flame  by greg rutkowski, high detail, realistic, concept art, trending on artstation. ",
        "scheduler": "K_EULER_ANCESTRAL",
        "noise_level": 0,
        "num_outputs": 1,
        "guidance_scale": 7.5,
        "negative_prompt": "amateur, mediocre, grainy, blur",
        "num_inference_steps": 50,
        "prior_guidance_scale": 4,
        "prior_num_inference_steps": 25
    }
)
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 akisakurai/stable-karlo 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": "akisakurai/stable-karlo:92a88275607a45c6b38503de522937131a43003586f5b0cb227e28f64d139af7",
    "input": {
      "width": 768,
      "height": 768,
      "prompt": "fancy portraits of panda holding vast in laboratory with neon light and blue flame  by greg rutkowski, high detail, realistic, concept art, trending on artstation. ",
      "scheduler": "K_EULER_ANCESTRAL",
      "noise_level": 0,
      "num_outputs": 1,
      "guidance_scale": 7.5,
      "negative_prompt": "amateur, mediocre, grainy, blur",
      "num_inference_steps": 50,
      "prior_guidance_scale": 4,
      "prior_num_inference_steps": 25
    }
  }' \
  https://api.replicate.com/v1/predictions

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

Output

{
  "completed_at": "2023-04-05T13:21:56.922212Z",
  "created_at": "2023-04-05T13:21:38.289428Z",
  "data_removed": false,
  "error": null,
  "id": "plnukmg2ibdpvbgwvvfsmwvfie",
  "input": {
    "width": 768,
    "height": 768,
    "prompt": "fancy portraits of panda holding vast in laboratory with neon light and blue flame  by greg rutkowski, high detail, realistic, concept art, trending on artstation. ",
    "scheduler": "K_EULER_ANCESTRAL",
    "num_outputs": 1,
    "guidance_scale": 7.5,
    "negative_prompt": "amateur, mediocre, grainy, blur",
    "num_inference_steps": 50
  },
  "logs": "Using seed: 58617\n  0%|          | 0/25 [00:00<?, ?it/s]\n 12%|█▏        | 3/25 [00:00<00:00, 22.71it/s]\n 28%|██▊       | 7/25 [00:00<00:00, 27.76it/s]\n 44%|████▍     | 11/25 [00:00<00:00, 31.66it/s]\n 60%|██████    | 15/25 [00:00<00:00, 33.04it/s]\n 76%|███████▌  | 19/25 [00:00<00:00, 34.19it/s]\n 92%|█████████▏| 23/25 [00:00<00:00, 35.04it/s]\n100%|██████████| 25/25 [00:00<00:00, 33.10it/s]\n  0%|          | 0/50 [00:00<?, ?it/s]\n  2%|▏         | 1/50 [00:00<00:14,  3.30it/s]\n  4%|▍         | 2/50 [00:00<00:14,  3.26it/s]\n  6%|▌         | 3/50 [00:00<00:14,  3.18it/s]\n  8%|▊         | 4/50 [00:01<00:14,  3.19it/s]\n 10%|█         | 5/50 [00:01<00:14,  3.20it/s]\n 12%|█▏        | 6/50 [00:01<00:13,  3.20it/s]\n 14%|█▍        | 7/50 [00:02<00:13,  3.18it/s]\n 16%|█▌        | 8/50 [00:02<00:13,  3.19it/s]\n 18%|█▊        | 9/50 [00:02<00:12,  3.19it/s]\n 20%|██        | 10/50 [00:03<00:12,  3.19it/s]\n 22%|██▏       | 11/50 [00:03<00:12,  3.18it/s]\n 24%|██▍       | 12/50 [00:03<00:11,  3.18it/s]\n 26%|██▌       | 13/50 [00:04<00:11,  3.18it/s]\n 28%|██▊       | 14/50 [00:04<00:11,  3.18it/s]\n 30%|███       | 15/50 [00:04<00:11,  3.18it/s]\n 32%|███▏      | 16/50 [00:05<00:10,  3.18it/s]\n 34%|███▍      | 17/50 [00:05<00:10,  3.17it/s]\n 36%|███▌      | 18/50 [00:05<00:10,  3.15it/s]\n 38%|███▊      | 19/50 [00:05<00:09,  3.15it/s]\n 40%|████      | 20/50 [00:06<00:09,  3.17it/s]\n 42%|████▏     | 21/50 [00:06<00:09,  3.17it/s]\n 44%|████▍     | 22/50 [00:06<00:08,  3.17it/s]\n 46%|████▌     | 23/50 [00:07<00:08,  3.16it/s]\n 48%|████▊     | 24/50 [00:07<00:08,  3.17it/s]\n 50%|█████     | 25/50 [00:07<00:07,  3.17it/s]\n 52%|█████▏    | 26/50 [00:08<00:07,  3.17it/s]\n 54%|█████▍    | 27/50 [00:08<00:07,  3.16it/s]\n 56%|█████▌    | 28/50 [00:08<00:06,  3.16it/s]\n 58%|█████▊    | 29/50 [00:09<00:06,  3.17it/s]\n 60%|██████    | 30/50 [00:09<00:06,  3.16it/s]\n 62%|██████▏   | 31/50 [00:09<00:06,  3.15it/s]\n 64%|██████▍   | 32/50 [00:10<00:05,  3.16it/s]\n 66%|██████▌   | 33/50 [00:10<00:05,  3.15it/s]\n 68%|██████▊   | 34/50 [00:10<00:05,  3.15it/s]\n 70%|███████   | 35/50 [00:11<00:04,  3.15it/s]\n 72%|███████▏  | 36/50 [00:11<00:04,  3.15it/s]\n 74%|███████▍  | 37/50 [00:11<00:04,  3.14it/s]\n 76%|███████▌  | 38/50 [00:11<00:03,  3.15it/s]\n 78%|███████▊  | 39/50 [00:12<00:03,  3.15it/s]\n 80%|████████  | 40/50 [00:12<00:03,  3.13it/s]\n 82%|████████▏ | 41/50 [00:12<00:02,  3.14it/s]\n 84%|████████▍ | 42/50 [00:13<00:02,  3.14it/s]\n 86%|████████▌ | 43/50 [00:13<00:02,  3.13it/s]\n 88%|████████▊ | 44/50 [00:13<00:01,  3.13it/s]\n 90%|█████████ | 45/50 [00:14<00:01,  3.15it/s]\n 92%|█████████▏| 46/50 [00:14<00:01,  3.15it/s]\n 94%|█████████▍| 47/50 [00:14<00:00,  3.14it/s]\n 96%|█████████▌| 48/50 [00:15<00:00,  3.15it/s]\n 98%|█████████▊| 49/50 [00:15<00:00,  3.14it/s]\n100%|██████████| 50/50 [00:15<00:00,  3.13it/s]\n100%|██████████| 50/50 [00:15<00:00,  3.16it/s]",
  "metrics": {
    "predict_time": 18.508965,
    "total_time": 18.632784
  },
  "output": [
    "https://replicate.delivery/pbxt/nFiRxljfCawWWKYSOYGD3nvdg9RRmz6BELZcRAmfbHZzfkdhA/out-0.png"
  ],
  "started_at": "2023-04-05T13:21:38.413247Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/plnukmg2ibdpvbgwvvfsmwvfie",
    "cancel": "https://api.replicate.com/v1/predictions/plnukmg2ibdpvbgwvvfsmwvfie/cancel"
  },
  "version": "bba407e2c36b98cd752488019ef092429f1575d91176634908b759d9ef537723"
}

Generated in

18.5 seconds

Tweak itReport View full prediction

Using seed: 58617
  0%|          | 0/25 [00:00<?, ?it/s]
 12%|█▏        | 3/25 [00:00<00:00, 22.71it/s]
 28%|██▊       | 7/25 [00:00<00:00, 27.76it/s]
 44%|████▍     | 11/25 [00:00<00:00, 31.66it/s]
 60%|██████    | 15/25 [00:00<00:00, 33.04it/s]
 76%|███████▌  | 19/25 [00:00<00:00, 34.19it/s]
 92%|█████████▏| 23/25 [00:00<00:00, 35.04it/s]
100%|██████████| 25/25 [00:00<00:00, 33.10it/s]
  0%|          | 0/50 [00:00<?, ?it/s]
  2%|▏         | 1/50 [00:00<00:14,  3.30it/s]
  4%|▍         | 2/50 [00:00<00:14,  3.26it/s]
  6%|▌         | 3/50 [00:00<00:14,  3.18it/s]
  8%|▊         | 4/50 [00:01<00:14,  3.19it/s]
 10%|█         | 5/50 [00:01<00:14,  3.20it/s]
 12%|█▏        | 6/50 [00:01<00:13,  3.20it/s]
 14%|█▍        | 7/50 [00:02<00:13,  3.18it/s]
 16%|█▌        | 8/50 [00:02<00:13,  3.19it/s]
 18%|█▊        | 9/50 [00:02<00:12,  3.19it/s]
 20%|██        | 10/50 [00:03<00:12,  3.19it/s]
 22%|██▏       | 11/50 [00:03<00:12,  3.18it/s]
 24%|██▍       | 12/50 [00:03<00:11,  3.18it/s]
 26%|██▌       | 13/50 [00:04<00:11,  3.18it/s]
 28%|██▊       | 14/50 [00:04<00:11,  3.18it/s]
 30%|███       | 15/50 [00:04<00:11,  3.18it/s]
 32%|███▏      | 16/50 [00:05<00:10,  3.18it/s]
 34%|███▍      | 17/50 [00:05<00:10,  3.17it/s]
 36%|███▌      | 18/50 [00:05<00:10,  3.15it/s]
 38%|███▊      | 19/50 [00:05<00:09,  3.15it/s]
 40%|████      | 20/50 [00:06<00:09,  3.17it/s]
 42%|████▏     | 21/50 [00:06<00:09,  3.17it/s]
 44%|████▍     | 22/50 [00:06<00:08,  3.17it/s]
 46%|████▌     | 23/50 [00:07<00:08,  3.16it/s]
 48%|████▊     | 24/50 [00:07<00:08,  3.17it/s]
 50%|█████     | 25/50 [00:07<00:07,  3.17it/s]
 52%|█████▏    | 26/50 [00:08<00:07,  3.17it/s]
 54%|█████▍    | 27/50 [00:08<00:07,  3.16it/s]
 56%|█████▌    | 28/50 [00:08<00:06,  3.16it/s]
 58%|█████▊    | 29/50 [00:09<00:06,  3.17it/s]
 60%|██████    | 30/50 [00:09<00:06,  3.16it/s]
 62%|██████▏   | 31/50 [00:09<00:06,  3.15it/s]
 64%|██████▍   | 32/50 [00:10<00:05,  3.16it/s]
 66%|██████▌   | 33/50 [00:10<00:05,  3.15it/s]
 68%|██████▊   | 34/50 [00:10<00:05,  3.15it/s]
 70%|███████   | 35/50 [00:11<00:04,  3.15it/s]
 72%|███████▏  | 36/50 [00:11<00:04,  3.15it/s]
 74%|███████▍  | 37/50 [00:11<00:04,  3.14it/s]
 76%|███████▌  | 38/50 [00:11<00:03,  3.15it/s]
 78%|███████▊  | 39/50 [00:12<00:03,  3.15it/s]
 80%|████████  | 40/50 [00:12<00:03,  3.13it/s]
 82%|████████▏ | 41/50 [00:12<00:02,  3.14it/s]
 84%|████████▍ | 42/50 [00:13<00:02,  3.14it/s]
 86%|████████▌ | 43/50 [00:13<00:02,  3.13it/s]
 88%|████████▊ | 44/50 [00:13<00:01,  3.13it/s]
 90%|█████████ | 45/50 [00:14<00:01,  3.15it/s]
 92%|█████████▏| 46/50 [00:14<00:01,  3.15it/s]
 94%|█████████▍| 47/50 [00:14<00:00,  3.14it/s]
 96%|█████████▌| 48/50 [00:15<00:00,  3.15it/s]
 98%|█████████▊| 49/50 [00:15<00:00,  3.14it/s]
100%|██████████| 50/50 [00:15<00:00,  3.13it/s]
100%|██████████| 50/50 [00:15<00:00,  3.16it/s]

This output was created using a different version of the model, akisakurai/stable-karlo:bba407e2.

Examples

View more examples

Run time and cost

This model costs approximately $0.0045 to run on Replicate, or 222 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 T4 GPU hardware. Predictions typically complete within 20 seconds. The predict time for this model varies significantly based on the inputs.

Readme

https://github.com/Stability-AI/stablediffusion/blob/main/doc/UNCLIP.MD