georgedavila/sdxl-bling – Run with an API on Replicate

georgedavila / sdxl-bling

SDXL LoRA finetuned on diamond watches

Cold

Public
53 runs
L40S
GitHub

Run with an API

Playground API Examples README Versions

Input

myprompt

string

Shift + Return to add a new line

a photo of a beach-themed watch, in the style of TOKa photo of a beach-themed watch, in the style of TOK

Input prompt (TOK is the style trigger word)

Default: "a photo of a gold watch, in the style of TOK"

promptAddendum

string

Shift + Return to add a new line

Extra terms to add to end of prompt

Default: ""

negative_prompt

string

Shift + Return to add a new line

Negative Prompt

Default: "deformed, blurry"

outWidth

integer

(minimum: 128, maximum: 4096)

width of output

Default: 1024

outHeight

integer

(minimum: 128, maximum: 4096)

height of output

Default: 1024

guidanceScale

number

(minimum: 0, maximum: 50)

Guidance scale (influence of input text on generation)

Default: 7.5

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

seed

integer

Random seed. Leave blank to randomize the seed

high_noise_frac

number

(minimum: 0, maximum: 1)

For expert_ensemble_refiner, the fraction of noise to use

Default: 0.8

refine_steps

integer

For base_image_refiner, the number of steps to refine, defaults to num_inference_steps

lora_scale

number

(minimum: 0, maximum: 1)

LoRA additive scale. Only applicable on trained models.

Default: 0.6

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

npx create-replicate --model=georgedavila/sdxl-bling

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 georgedavila/sdxl-bling using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "georgedavila/sdxl-bling:26b301ef3d371f475185eb1a420ae156aa0836971557123475cbb9bb352d55d6",
  {
    input: {
      myprompt: "a photo of a beach-themed watch, in the style of TOK",
      outWidth: 1024,
      outHeight: 1024,
      lora_scale: 0.6,
      num_outputs: 1,
      guidanceScale: 7.5,
      promptAddendum: "",
      high_noise_frac: 0.8,
      negative_prompt: "deformed, blurry, (close up)",
      num_inference_steps: 50
    }
  }
);

// 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 georgedavila/sdxl-bling using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "georgedavila/sdxl-bling:26b301ef3d371f475185eb1a420ae156aa0836971557123475cbb9bb352d55d6",
    input={
        "myprompt": "a photo of a beach-themed watch, in the style of TOK",
        "outWidth": 1024,
        "outHeight": 1024,
        "lora_scale": 0.6,
        "num_outputs": 1,
        "guidanceScale": 7.5,
        "promptAddendum": "",
        "high_noise_frac": 0.8,
        "negative_prompt": "deformed, blurry, (close up)",
        "num_inference_steps": 50
    }
)
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 georgedavila/sdxl-bling 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": "georgedavila/sdxl-bling:26b301ef3d371f475185eb1a420ae156aa0836971557123475cbb9bb352d55d6",
    "input": {
      "myprompt": "a photo of a beach-themed watch, in the style of TOK",
      "outWidth": 1024,
      "outHeight": 1024,
      "lora_scale": 0.6,
      "num_outputs": 1,
      "guidanceScale": 7.5,
      "promptAddendum": "",
      "high_noise_frac": 0.8,
      "negative_prompt": "deformed, blurry, (close up)",
      "num_inference_steps": 50
    }
  }' \
  https://api.replicate.com/v1/predictions

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

Output

{
  "completed_at": "2024-01-07T22:43:07.532493Z",
  "created_at": "2024-01-07T22:42:53.814341Z",
  "data_removed": false,
  "error": null,
  "id": "hpvtyetbfjwocfoyh5hetcthzq",
  "input": {
    "myprompt": "a photo of a beach-themed watch, in the style of TOK",
    "outWidth": 1024,
    "outHeight": 1024,
    "lora_scale": 0.6,
    "num_outputs": 1,
    "guidanceScale": 7.5,
    "promptAddendum": "",
    "high_noise_frac": 0.8,
    "negative_prompt": "deformed, blurry, (close up)",
    "num_inference_steps": 50
  },
  "logs": "Using seed: 27757\n  0%|          | 0/50 [00:00<?, ?it/s]\n  2%|▏         | 1/50 [00:00<00:11,  4.30it/s]\n  4%|▍         | 2/50 [00:00<00:11,  4.28it/s]\n  6%|▌         | 3/50 [00:00<00:11,  4.27it/s]\n  8%|▊         | 4/50 [00:00<00:10,  4.25it/s]\n 10%|█         | 5/50 [00:01<00:10,  4.25it/s]\n 12%|█▏        | 6/50 [00:01<00:10,  4.24it/s]\n 14%|█▍        | 7/50 [00:01<00:10,  4.24it/s]\n 16%|█▌        | 8/50 [00:01<00:09,  4.24it/s]\n 18%|█▊        | 9/50 [00:02<00:09,  4.24it/s]\n 20%|██        | 10/50 [00:02<00:09,  4.23it/s]\n 22%|██▏       | 11/50 [00:02<00:09,  4.23it/s]\n 24%|██▍       | 12/50 [00:02<00:08,  4.23it/s]\n 26%|██▌       | 13/50 [00:03<00:08,  4.23it/s]\n 28%|██▊       | 14/50 [00:03<00:08,  4.23it/s]\n 30%|███       | 15/50 [00:03<00:08,  4.23it/s]\n 32%|███▏      | 16/50 [00:03<00:08,  4.23it/s]\n 34%|███▍      | 17/50 [00:04<00:07,  4.23it/s]\n 36%|███▌      | 18/50 [00:04<00:07,  4.23it/s]\n 38%|███▊      | 19/50 [00:04<00:07,  4.23it/s]\n 40%|████      | 20/50 [00:04<00:07,  4.23it/s]\n 42%|████▏     | 21/50 [00:04<00:06,  4.22it/s]\n 44%|████▍     | 22/50 [00:05<00:06,  4.22it/s]\n 46%|████▌     | 23/50 [00:05<00:06,  4.22it/s]\n 48%|████▊     | 24/50 [00:05<00:06,  4.22it/s]\n 50%|█████     | 25/50 [00:05<00:05,  4.22it/s]\n 52%|█████▏    | 26/50 [00:06<00:05,  4.23it/s]\n 54%|█████▍    | 27/50 [00:06<00:05,  4.23it/s]\n 56%|█████▌    | 28/50 [00:06<00:05,  4.24it/s]\n 58%|█████▊    | 29/50 [00:06<00:04,  4.25it/s]\n 60%|██████    | 30/50 [00:07<00:04,  4.25it/s]\n 62%|██████▏   | 31/50 [00:07<00:04,  4.25it/s]\n 64%|██████▍   | 32/50 [00:07<00:04,  4.25it/s]\n 66%|██████▌   | 33/50 [00:07<00:04,  4.25it/s]\n 68%|██████▊   | 34/50 [00:08<00:03,  4.25it/s]\n 70%|███████   | 35/50 [00:08<00:03,  4.25it/s]\n 72%|███████▏  | 36/50 [00:08<00:03,  4.25it/s]\n 74%|███████▍  | 37/50 [00:08<00:03,  4.25it/s]\n 76%|███████▌  | 38/50 [00:08<00:02,  4.25it/s]\n 78%|███████▊  | 39/50 [00:09<00:02,  4.25it/s]\n 80%|████████  | 40/50 [00:09<00:02,  4.24it/s]\n 82%|████████▏ | 41/50 [00:09<00:02,  4.25it/s]\n 84%|████████▍ | 42/50 [00:09<00:01,  4.24it/s]\n 86%|████████▌ | 43/50 [00:10<00:01,  4.25it/s]\n 88%|████████▊ | 44/50 [00:10<00:01,  4.25it/s]\n 90%|█████████ | 45/50 [00:10<00:01,  4.25it/s]\n 92%|█████████▏| 46/50 [00:10<00:00,  4.25it/s]\n 94%|█████████▍| 47/50 [00:11<00:00,  4.24it/s]\n 96%|█████████▌| 48/50 [00:11<00:00,  4.24it/s]\n 98%|█████████▊| 49/50 [00:11<00:00,  4.24it/s]\n100%|██████████| 50/50 [00:11<00:00,  4.24it/s]\n100%|██████████| 50/50 [00:11<00:00,  4.24it/s]",
  "metrics": {
    "predict_time": 13.659626,
    "total_time": 13.718152
  },
  "output": [
    "https://replicate.delivery/pbxt/MzK4v7kuWTKnGpGy62yhM9IecL4q2eJ8d9KLTKfkE560XbUkA/out-0.png"
  ],
  "started_at": "2024-01-07T22:42:53.872867Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/hpvtyetbfjwocfoyh5hetcthzq",
    "cancel": "https://api.replicate.com/v1/predictions/hpvtyetbfjwocfoyh5hetcthzq/cancel"
  },
  "version": "26b301ef3d371f475185eb1a420ae156aa0836971557123475cbb9bb352d55d6"
}

Generated in

13.7 seconds

Tweak itReport

Using seed: 27757
  0%|          | 0/50 [00:00<?, ?it/s]
  2%|▏         | 1/50 [00:00<00:11,  4.30it/s]
  4%|▍         | 2/50 [00:00<00:11,  4.28it/s]
  6%|▌         | 3/50 [00:00<00:11,  4.27it/s]
  8%|▊         | 4/50 [00:00<00:10,  4.25it/s]
 10%|█         | 5/50 [00:01<00:10,  4.25it/s]
 12%|█▏        | 6/50 [00:01<00:10,  4.24it/s]
 14%|█▍        | 7/50 [00:01<00:10,  4.24it/s]
 16%|█▌        | 8/50 [00:01<00:09,  4.24it/s]
 18%|█▊        | 9/50 [00:02<00:09,  4.24it/s]
 20%|██        | 10/50 [00:02<00:09,  4.23it/s]
 22%|██▏       | 11/50 [00:02<00:09,  4.23it/s]
 24%|██▍       | 12/50 [00:02<00:08,  4.23it/s]
 26%|██▌       | 13/50 [00:03<00:08,  4.23it/s]
 28%|██▊       | 14/50 [00:03<00:08,  4.23it/s]
 30%|███       | 15/50 [00:03<00:08,  4.23it/s]
 32%|███▏      | 16/50 [00:03<00:08,  4.23it/s]
 34%|███▍      | 17/50 [00:04<00:07,  4.23it/s]
 36%|███▌      | 18/50 [00:04<00:07,  4.23it/s]
 38%|███▊      | 19/50 [00:04<00:07,  4.23it/s]
 40%|████      | 20/50 [00:04<00:07,  4.23it/s]
 42%|████▏     | 21/50 [00:04<00:06,  4.22it/s]
 44%|████▍     | 22/50 [00:05<00:06,  4.22it/s]
 46%|████▌     | 23/50 [00:05<00:06,  4.22it/s]
 48%|████▊     | 24/50 [00:05<00:06,  4.22it/s]
 50%|█████     | 25/50 [00:05<00:05,  4.22it/s]
 52%|█████▏    | 26/50 [00:06<00:05,  4.23it/s]
 54%|█████▍    | 27/50 [00:06<00:05,  4.23it/s]
 56%|█████▌    | 28/50 [00:06<00:05,  4.24it/s]
 58%|█████▊    | 29/50 [00:06<00:04,  4.25it/s]
 60%|██████    | 30/50 [00:07<00:04,  4.25it/s]
 62%|██████▏   | 31/50 [00:07<00:04,  4.25it/s]
 64%|██████▍   | 32/50 [00:07<00:04,  4.25it/s]
 66%|██████▌   | 33/50 [00:07<00:04,  4.25it/s]
 68%|██████▊   | 34/50 [00:08<00:03,  4.25it/s]
 70%|███████   | 35/50 [00:08<00:03,  4.25it/s]
 72%|███████▏  | 36/50 [00:08<00:03,  4.25it/s]
 74%|███████▍  | 37/50 [00:08<00:03,  4.25it/s]
 76%|███████▌  | 38/50 [00:08<00:02,  4.25it/s]
 78%|███████▊  | 39/50 [00:09<00:02,  4.25it/s]
 80%|████████  | 40/50 [00:09<00:02,  4.24it/s]
 82%|████████▏ | 41/50 [00:09<00:02,  4.25it/s]
 84%|████████▍ | 42/50 [00:09<00:01,  4.24it/s]
 86%|████████▌ | 43/50 [00:10<00:01,  4.25it/s]
 88%|████████▊ | 44/50 [00:10<00:01,  4.25it/s]
 90%|█████████ | 45/50 [00:10<00:01,  4.25it/s]
 92%|█████████▏| 46/50 [00:10<00:00,  4.25it/s]
 94%|█████████▍| 47/50 [00:11<00:00,  4.24it/s]
 96%|█████████▌| 48/50 [00:11<00:00,  4.24it/s]
 98%|█████████▊| 49/50 [00:11<00:00,  4.24it/s]
100%|██████████| 50/50 [00:11<00:00,  4.24it/s]
100%|██████████| 50/50 [00:11<00:00,  4.24it/s]

Examples

View more examples

Run time and cost

This model costs approximately $0.039 to run on Replicate, or 25 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 40 seconds. The predict time for this model varies significantly based on the inputs.

Readme

Model: https://huggingface.co/GDavila/sdxl-bling Dataset: https://huggingface.co/datasets/GDavila/diamondWatches