cjwbw / compositional-vsual-generation-with-composable-diffusion-models-pytorch

Composable Diffusion

Cold

Public
846 runs
T4
GitHub
Paper
License

Run with an API

Playground API Examples README Versions

Input

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

npx create-replicate --model=cjwbw/compositional-vsual-generation-with-composable-diffusion-models-pytorch

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 cjwbw/compositional-vsual-generation-with-composable-diffusion-models-pytorch using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "cjwbw/compositional-vsual-generation-with-composable-diffusion-models-pytorch:9093eef2cad6a0445f108067a38e1a17e22f8c8073d9f7d60d95cf354def4d29",
  {
    input: {
      prompt: "A red car parked in a desert  | Hills behind the car | Aurora in the sky"
    }
  }
);

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

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

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 cjwbw/compositional-vsual-generation-with-composable-diffusion-models-pytorch using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "cjwbw/compositional-vsual-generation-with-composable-diffusion-models-pytorch:9093eef2cad6a0445f108067a38e1a17e22f8c8073d9f7d60d95cf354def4d29",
    input={
        "prompt": "A red car parked in a desert  | Hills behind the car | Aurora in the sky"
    }
)
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 cjwbw/compositional-vsual-generation-with-composable-diffusion-models-pytorch 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": "9093eef2cad6a0445f108067a38e1a17e22f8c8073d9f7d60d95cf354def4d29",
    "input": {
      "prompt": "A red car parked in a desert  | Hills behind the car | Aurora in the sky"
    }
  }' \
  https://api.replicate.com/v1/predictions

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

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/chenxwh/compositional-vsual-generation-with-composable-diffusion-models-pytorch@sha256:9093eef2cad6a0445f108067a38e1a17e22f8c8073d9f7d60d95cf354def4d29 \
  -i 'prompt="A red car parked in a desert  | Hills behind the car | Aurora in the sky"'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/chenxwh/compositional-vsual-generation-with-composable-diffusion-models-pytorch@sha256:9093eef2cad6a0445f108067a38e1a17e22f8c8073d9f7d60d95cf354def4d29
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "prompt": "A red car parked in a desert  | Hills behind the car | Aurora in the sky"
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

{
  "completed_at": "2022-08-04T18:05:56.206371Z",
  "created_at": "2022-08-04T18:05:36.641735Z",
  "data_removed": false,
  "error": null,
  "id": "sbtd2l4mnbh63hwdrz4ygfkfve",
  "input": {
    "prompt": "A red car parked in a desert  | Hills behind the car | Aurora in the sky"
  },
  "logs": "\n  0%|          | 0/100 [00:00<?, ?it/s]\n  1%|          | 1/100 [00:00<00:20,  4.85it/s]\n  2%|▏         | 2/100 [00:00<00:18,  5.36it/s]\n  3%|▎         | 3/100 [00:00<00:16,  5.79it/s]\n  4%|▍         | 4/100 [00:00<00:15,  6.14it/s]\n  5%|▌         | 5/100 [00:00<00:14,  6.41it/s]\n  6%|▌         | 6/100 [00:00<00:14,  6.63it/s]\n  7%|▋         | 7/100 [00:01<00:13,  6.78it/s]\n  8%|▊         | 8/100 [00:01<00:13,  6.88it/s]\n  9%|▉         | 9/100 [00:01<00:13,  6.95it/s]\n 10%|█         | 10/100 [00:01<00:12,  7.01it/s]\n 11%|█         | 11/100 [00:01<00:12,  7.04it/s]\n 12%|█▏        | 12/100 [00:01<00:12,  7.07it/s]\n 13%|█▎        | 13/100 [00:01<00:12,  7.11it/s]\n 14%|█▍        | 14/100 [00:02<00:12,  7.10it/s]\n 15%|█▌        | 15/100 [00:02<00:11,  7.09it/s]\n 16%|█▌        | 16/100 [00:02<00:11,  7.10it/s]\n 17%|█▋        | 17/100 [00:02<00:11,  7.12it/s]\n 18%|█▊        | 18/100 [00:02<00:11,  7.13it/s]\n 19%|█▉        | 19/100 [00:02<00:11,  7.15it/s]\n 20%|██        | 20/100 [00:02<00:11,  7.15it/s]\n 21%|██        | 21/100 [00:03<00:11,  7.15it/s]\n 22%|██▏       | 22/100 [00:03<00:10,  7.15it/s]\n 23%|██▎       | 23/100 [00:03<00:10,  7.13it/s]\n 24%|██▍       | 24/100 [00:03<00:10,  7.16it/s]\n 25%|██▌       | 25/100 [00:03<00:10,  7.17it/s]\n 26%|██▌       | 26/100 [00:03<00:10,  7.17it/s]\n 27%|██▋       | 27/100 [00:03<00:10,  7.17it/s]\n 28%|██▊       | 28/100 [00:03<00:10,  7.15it/s]\n 29%|██▉       | 29/100 [00:04<00:09,  7.13it/s]\n 30%|███       | 30/100 [00:04<00:09,  7.14it/s]\n 31%|███       | 31/100 [00:04<00:09,  7.13it/s]\n 32%|███▏      | 32/100 [00:04<00:09,  7.14it/s]\n 33%|███▎      | 33/100 [00:04<00:09,  7.14it/s]\n 34%|███▍      | 34/100 [00:04<00:09,  7.14it/s]\n 35%|███▌      | 35/100 [00:04<00:09,  7.15it/s]\n 36%|███▌      | 36/100 [00:05<00:08,  7.16it/s]\n 37%|███▋      | 37/100 [00:05<00:08,  7.15it/s]\n 38%|███▊      | 38/100 [00:05<00:08,  7.13it/s]\n 39%|███▉      | 39/100 [00:05<00:08,  7.16it/s]\n 40%|████      | 40/100 [00:05<00:08,  7.14it/s]\n 41%|████      | 41/100 [00:05<00:08,  7.12it/s]\n 42%|████▏     | 42/100 [00:05<00:08,  7.11it/s]\n 43%|████▎     | 43/100 [00:06<00:07,  7.13it/s]\n 44%|████▍     | 44/100 [00:06<00:07,  7.13it/s]\n 45%|████▌     | 45/100 [00:06<00:07,  7.12it/s]\n 46%|████▌     | 46/100 [00:06<00:07,  7.13it/s]\n 47%|████▋     | 47/100 [00:06<00:07,  7.15it/s]\n 48%|████▊     | 48/100 [00:06<00:07,  7.15it/s]\n 49%|████▉     | 49/100 [00:06<00:07,  7.13it/s]\n 50%|█████     | 50/100 [00:07<00:07,  7.12it/s]\n 51%|█████     | 51/100 [00:07<00:06,  7.09it/s]\n 52%|█████▏    | 52/100 [00:07<00:06,  7.09it/s]\n 53%|█████▎    | 53/100 [00:07<00:06,  7.09it/s]\n 54%|█████▍    | 54/100 [00:07<00:06,  7.10it/s]\n 55%|█████▌    | 55/100 [00:07<00:06,  7.11it/s]\n 56%|█████▌    | 56/100 [00:07<00:06,  7.09it/s]\n 57%|█████▋    | 57/100 [00:08<00:06,  7.11it/s]\n 58%|█████▊    | 58/100 [00:08<00:05,  7.12it/s]\n 59%|█████▉    | 59/100 [00:08<00:05,  7.12it/s]\n 60%|██████    | 60/100 [00:08<00:05,  7.12it/s]\n 61%|██████    | 61/100 [00:08<00:05,  7.11it/s]\n 62%|██████▏   | 62/100 [00:08<00:05,  7.10it/s]\n 63%|██████▎   | 63/100 [00:08<00:05,  7.09it/s]\n 64%|██████▍   | 64/100 [00:09<00:05,  7.07it/s]\n 65%|██████▌   | 65/100 [00:09<00:04,  7.06it/s]\n 66%|██████▌   | 66/100 [00:09<00:04,  7.06it/s]\n 67%|██████▋   | 67/100 [00:09<00:04,  7.05it/s]\n 68%|██████▊   | 68/100 [00:09<00:04,  7.06it/s]\n 69%|██████▉   | 69/100 [00:09<00:04,  7.09it/s]\n 70%|███████   | 70/100 [00:09<00:04,  7.06it/s]\n 71%|███████   | 71/100 [00:10<00:04,  7.07it/s]\n 72%|███████▏  | 72/100 [00:10<00:03,  7.08it/s]\n 73%|███████▎  | 73/100 [00:10<00:03,  7.02it/s]\n 74%|███████▍  | 74/100 [00:10<00:03,  7.01it/s]\n 75%|███████▌  | 75/100 [00:10<00:03,  7.03it/s]\n 76%|███████▌  | 76/100 [00:10<00:03,  7.03it/s]\n 77%|███████▋  | 77/100 [00:10<00:03,  7.04it/s]\n 78%|███████▊  | 78/100 [00:11<00:03,  7.04it/s]\n 79%|███████▉  | 79/100 [00:11<00:02,  7.06it/s]\n 80%|████████  | 80/100 [00:11<00:02,  7.04it/s]\n 81%|████████  | 81/100 [00:11<00:02,  7.07it/s]\n 82%|████████▏ | 82/100 [00:11<00:02,  7.09it/s]\n 83%|████████▎ | 83/100 [00:11<00:02,  7.08it/s]\n 84%|████████▍ | 84/100 [00:11<00:02,  7.09it/s]\n 85%|████████▌ | 85/100 [00:12<00:02,  7.10it/s]\n 86%|████████▌ | 86/100 [00:12<00:01,  7.09it/s]\n 87%|████████▋ | 87/100 [00:12<00:01,  7.08it/s]\n 88%|████████▊ | 88/100 [00:12<00:01,  7.09it/s]\n 89%|████████▉ | 89/100 [00:12<00:01,  7.08it/s]\n 90%|█████████ | 90/100 [00:12<00:01,  7.08it/s]\n 91%|█████████ | 91/100 [00:12<00:01,  7.06it/s]\n 92%|█████████▏| 92/100 [00:13<00:01,  7.08it/s]\n 93%|█████████▎| 93/100 [00:13<00:00,  7.06it/s]\n 94%|█████████▍| 94/100 [00:13<00:00,  7.05it/s]\n 95%|█████████▌| 95/100 [00:13<00:00,  7.04it/s]\n 96%|█████████▌| 96/100 [00:13<00:00,  7.02it/s]\n 97%|█████████▋| 97/100 [00:13<00:00,  7.01it/s]\n 98%|█████████▊| 98/100 [00:13<00:00,  7.02it/s]\n 99%|█████████▉| 99/100 [00:14<00:00,  7.02it/s]\n100%|██████████| 100/100 [00:14<00:00,  7.02it/s]\n100%|██████████| 100/100 [00:14<00:00,  7.07it/s]\n\n  0%|          | 0/27 [00:00<?, ?it/s]\n  4%|▎         | 1/27 [00:00<00:04,  5.88it/s]\n  7%|▋         | 2/27 [00:00<00:04,  5.87it/s]\n 11%|█         | 3/27 [00:00<00:04,  5.86it/s]\n 15%|█▍        | 4/27 [00:00<00:03,  5.85it/s]\n 19%|█▊        | 5/27 [00:00<00:03,  5.84it/s]\n 22%|██▏       | 6/27 [00:01<00:03,  5.83it/s]\n 26%|██▌       | 7/27 [00:01<00:03,  5.82it/s]\n 30%|██▉       | 8/27 [00:01<00:03,  5.82it/s]\n 33%|███▎      | 9/27 [00:01<00:03,  5.82it/s]\n 37%|███▋      | 10/27 [00:01<00:02,  5.82it/s]\n 41%|████      | 11/27 [00:01<00:02,  5.82it/s]\n 44%|████▍     | 12/27 [00:02<00:02,  5.83it/s]\n 48%|████▊     | 13/27 [00:02<00:02,  5.83it/s]\n 52%|█████▏    | 14/27 [00:02<00:02,  5.84it/s]\n 56%|█████▌    | 15/27 [00:02<00:02,  5.85it/s]\n 59%|█████▉    | 16/27 [00:02<00:01,  5.84it/s]\n 63%|██████▎   | 17/27 [00:02<00:01,  5.86it/s]\n 67%|██████▋   | 18/27 [00:03<00:01,  5.85it/s]\n 70%|███████   | 19/27 [00:03<00:01,  5.86it/s]\n 74%|███████▍  | 20/27 [00:03<00:01,  5.86it/s]\n 78%|███████▊  | 21/27 [00:03<00:01,  5.86it/s]\n 81%|████████▏ | 22/27 [00:03<00:00,  5.86it/s]\n 85%|████████▌ | 23/27 [00:03<00:00,  5.85it/s]\n 89%|████████▉ | 24/27 [00:04<00:00,  5.87it/s]\n 93%|█████████▎| 25/27 [00:04<00:00,  5.85it/s]\n 96%|█████████▋| 26/27 [00:04<00:00,  5.86it/s]\n100%|██████████| 27/27 [00:04<00:00,  5.83it/s]\n100%|██████████| 27/27 [00:04<00:00,  5.84it/s]",
  "metrics": {
    "predict_time": 19.391179,
    "total_time": 19.564636
  },
  "output": "https://replicate.delivery/mgxm/9de5cffc-da99-43d0-9391-7ef4e9d4be1e/output.png",
  "started_at": "2022-08-04T18:05:36.815192Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/sbtd2l4mnbh63hwdrz4ygfkfve",
    "cancel": "https://api.replicate.com/v1/predictions/sbtd2l4mnbh63hwdrz4ygfkfve/cancel"
  },
  "version": "9093eef2cad6a0445f108067a38e1a17e22f8c8073d9f7d60d95cf354def4d29"
}

Generated in

19.4 seconds

Tweak itReport


  0%|          | 0/100 [00:00<?, ?it/s]
  1%|          | 1/100 [00:00<00:20,  4.85it/s]
  2%|▏         | 2/100 [00:00<00:18,  5.36it/s]
  3%|▎         | 3/100 [00:00<00:16,  5.79it/s]
  4%|▍         | 4/100 [00:00<00:15,  6.14it/s]
  5%|▌         | 5/100 [00:00<00:14,  6.41it/s]
  6%|▌         | 6/100 [00:00<00:14,  6.63it/s]
  7%|▋         | 7/100 [00:01<00:13,  6.78it/s]
  8%|▊         | 8/100 [00:01<00:13,  6.88it/s]
  9%|▉         | 9/100 [00:01<00:13,  6.95it/s]
 10%|█         | 10/100 [00:01<00:12,  7.01it/s]
 11%|█         | 11/100 [00:01<00:12,  7.04it/s]
 12%|█▏        | 12/100 [00:01<00:12,  7.07it/s]
 13%|█▎        | 13/100 [00:01<00:12,  7.11it/s]
 14%|█▍        | 14/100 [00:02<00:12,  7.10it/s]
 15%|█▌        | 15/100 [00:02<00:11,  7.09it/s]
 16%|█▌        | 16/100 [00:02<00:11,  7.10it/s]
 17%|█▋        | 17/100 [00:02<00:11,  7.12it/s]
 18%|█▊        | 18/100 [00:02<00:11,  7.13it/s]
 19%|█▉        | 19/100 [00:02<00:11,  7.15it/s]
 20%|██        | 20/100 [00:02<00:11,  7.15it/s]
 21%|██        | 21/100 [00:03<00:11,  7.15it/s]
 22%|██▏       | 22/100 [00:03<00:10,  7.15it/s]
 23%|██▎       | 23/100 [00:03<00:10,  7.13it/s]
 24%|██▍       | 24/100 [00:03<00:10,  7.16it/s]
 25%|██▌       | 25/100 [00:03<00:10,  7.17it/s]
 26%|██▌       | 26/100 [00:03<00:10,  7.17it/s]
 27%|██▋       | 27/100 [00:03<00:10,  7.17it/s]
 28%|██▊       | 28/100 [00:03<00:10,  7.15it/s]
 29%|██▉       | 29/100 [00:04<00:09,  7.13it/s]
 30%|███       | 30/100 [00:04<00:09,  7.14it/s]
 31%|███       | 31/100 [00:04<00:09,  7.13it/s]
 32%|███▏      | 32/100 [00:04<00:09,  7.14it/s]
 33%|███▎      | 33/100 [00:04<00:09,  7.14it/s]
 34%|███▍      | 34/100 [00:04<00:09,  7.14it/s]
 35%|███▌      | 35/100 [00:04<00:09,  7.15it/s]
 36%|███▌      | 36/100 [00:05<00:08,  7.16it/s]
 37%|███▋      | 37/100 [00:05<00:08,  7.15it/s]
 38%|███▊      | 38/100 [00:05<00:08,  7.13it/s]
 39%|███▉      | 39/100 [00:05<00:08,  7.16it/s]
 40%|████      | 40/100 [00:05<00:08,  7.14it/s]
 41%|████      | 41/100 [00:05<00:08,  7.12it/s]
 42%|████▏     | 42/100 [00:05<00:08,  7.11it/s]
 43%|████▎     | 43/100 [00:06<00:07,  7.13it/s]
 44%|████▍     | 44/100 [00:06<00:07,  7.13it/s]
 45%|████▌     | 45/100 [00:06<00:07,  7.12it/s]
 46%|████▌     | 46/100 [00:06<00:07,  7.13it/s]
 47%|████▋     | 47/100 [00:06<00:07,  7.15it/s]
 48%|████▊     | 48/100 [00:06<00:07,  7.15it/s]
 49%|████▉     | 49/100 [00:06<00:07,  7.13it/s]
 50%|█████     | 50/100 [00:07<00:07,  7.12it/s]
 51%|█████     | 51/100 [00:07<00:06,  7.09it/s]
 52%|█████▏    | 52/100 [00:07<00:06,  7.09it/s]
 53%|█████▎    | 53/100 [00:07<00:06,  7.09it/s]
 54%|█████▍    | 54/100 [00:07<00:06,  7.10it/s]
 55%|█████▌    | 55/100 [00:07<00:06,  7.11it/s]
 56%|█████▌    | 56/100 [00:07<00:06,  7.09it/s]
 57%|█████▋    | 57/100 [00:08<00:06,  7.11it/s]
 58%|█████▊    | 58/100 [00:08<00:05,  7.12it/s]
 59%|█████▉    | 59/100 [00:08<00:05,  7.12it/s]
 60%|██████    | 60/100 [00:08<00:05,  7.12it/s]
 61%|██████    | 61/100 [00:08<00:05,  7.11it/s]
 62%|██████▏   | 62/100 [00:08<00:05,  7.10it/s]
 63%|██████▎   | 63/100 [00:08<00:05,  7.09it/s]
 64%|██████▍   | 64/100 [00:09<00:05,  7.07it/s]
 65%|██████▌   | 65/100 [00:09<00:04,  7.06it/s]
 66%|██████▌   | 66/100 [00:09<00:04,  7.06it/s]
 67%|██████▋   | 67/100 [00:09<00:04,  7.05it/s]
 68%|██████▊   | 68/100 [00:09<00:04,  7.06it/s]
 69%|██████▉   | 69/100 [00:09<00:04,  7.09it/s]
 70%|███████   | 70/100 [00:09<00:04,  7.06it/s]
 71%|███████   | 71/100 [00:10<00:04,  7.07it/s]
 72%|███████▏  | 72/100 [00:10<00:03,  7.08it/s]
 73%|███████▎  | 73/100 [00:10<00:03,  7.02it/s]
 74%|███████▍  | 74/100 [00:10<00:03,  7.01it/s]
 75%|███████▌  | 75/100 [00:10<00:03,  7.03it/s]
 76%|███████▌  | 76/100 [00:10<00:03,  7.03it/s]
 77%|███████▋  | 77/100 [00:10<00:03,  7.04it/s]
 78%|███████▊  | 78/100 [00:11<00:03,  7.04it/s]
 79%|███████▉  | 79/100 [00:11<00:02,  7.06it/s]
 80%|████████  | 80/100 [00:11<00:02,  7.04it/s]
 81%|████████  | 81/100 [00:11<00:02,  7.07it/s]
 82%|████████▏ | 82/100 [00:11<00:02,  7.09it/s]
 83%|████████▎ | 83/100 [00:11<00:02,  7.08it/s]
 84%|████████▍ | 84/100 [00:11<00:02,  7.09it/s]
 85%|████████▌ | 85/100 [00:12<00:02,  7.10it/s]
 86%|████████▌ | 86/100 [00:12<00:01,  7.09it/s]
 87%|████████▋ | 87/100 [00:12<00:01,  7.08it/s]
 88%|████████▊ | 88/100 [00:12<00:01,  7.09it/s]
 89%|████████▉ | 89/100 [00:12<00:01,  7.08it/s]
 90%|█████████ | 90/100 [00:12<00:01,  7.08it/s]
 91%|█████████ | 91/100 [00:12<00:01,  7.06it/s]
 92%|█████████▏| 92/100 [00:13<00:01,  7.08it/s]
 93%|█████████▎| 93/100 [00:13<00:00,  7.06it/s]
 94%|█████████▍| 94/100 [00:13<00:00,  7.05it/s]
 95%|█████████▌| 95/100 [00:13<00:00,  7.04it/s]
 96%|█████████▌| 96/100 [00:13<00:00,  7.02it/s]
 97%|█████████▋| 97/100 [00:13<00:00,  7.01it/s]
 98%|█████████▊| 98/100 [00:13<00:00,  7.02it/s]
 99%|█████████▉| 99/100 [00:14<00:00,  7.02it/s]
100%|██████████| 100/100 [00:14<00:00,  7.02it/s]
100%|██████████| 100/100 [00:14<00:00,  7.07it/s]

  0%|          | 0/27 [00:00<?, ?it/s]
  4%|▎         | 1/27 [00:00<00:04,  5.88it/s]
  7%|▋         | 2/27 [00:00<00:04,  5.87it/s]
 11%|█         | 3/27 [00:00<00:04,  5.86it/s]
 15%|█▍        | 4/27 [00:00<00:03,  5.85it/s]
 19%|█▊        | 5/27 [00:00<00:03,  5.84it/s]
 22%|██▏       | 6/27 [00:01<00:03,  5.83it/s]
 26%|██▌       | 7/27 [00:01<00:03,  5.82it/s]
 30%|██▉       | 8/27 [00:01<00:03,  5.82it/s]
 33%|███▎      | 9/27 [00:01<00:03,  5.82it/s]
 37%|███▋      | 10/27 [00:01<00:02,  5.82it/s]
 41%|████      | 11/27 [00:01<00:02,  5.82it/s]
 44%|████▍     | 12/27 [00:02<00:02,  5.83it/s]
 48%|████▊     | 13/27 [00:02<00:02,  5.83it/s]
 52%|█████▏    | 14/27 [00:02<00:02,  5.84it/s]
 56%|█████▌    | 15/27 [00:02<00:02,  5.85it/s]
 59%|█████▉    | 16/27 [00:02<00:01,  5.84it/s]
 63%|██████▎   | 17/27 [00:02<00:01,  5.86it/s]
 67%|██████▋   | 18/27 [00:03<00:01,  5.85it/s]
 70%|███████   | 19/27 [00:03<00:01,  5.86it/s]
 74%|███████▍  | 20/27 [00:03<00:01,  5.86it/s]
 78%|███████▊  | 21/27 [00:03<00:01,  5.86it/s]
 81%|████████▏ | 22/27 [00:03<00:00,  5.86it/s]
 85%|████████▌ | 23/27 [00:03<00:00,  5.85it/s]
 89%|████████▉ | 24/27 [00:04<00:00,  5.87it/s]
 93%|█████████▎| 25/27 [00:04<00:00,  5.85it/s]
 96%|█████████▋| 26/27 [00:04<00:00,  5.86it/s]
100%|██████████| 27/27 [00:04<00:00,  5.83it/s]
100%|██████████| 27/27 [00:04<00:00,  5.84it/s]

Examples

View more examples

Run time and cost

This model costs approximately $0.0047 to run on Replicate, or 212 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 21 seconds. The predict time for this model varies significantly based on the inputs.

Readme

This is a cog implementation of https://github.com/energy-based-model/Compositional-Visual-Generation-with-Composable-Diffusion-Models-PyTorch

Composable Diffusion

Project Page

This is the official codebase for Compositional Visual Generation with Composable Diffusion Models.

Compositional Visual Generation with Composable Diffusion Models
Nan Liu Shuang Li Yilun Du Antonio Torralba Joshua B. Tenenbaum

The codebase is built upon GLIDE.

Citing our Paper

If you find our code useful for your research, please consider citing

@article{liu2022compositional,
  title={Compositional Visual Generation with Composable Diffusion Models},
  author={Liu, Nan and Li, Shuang and Du, Yilun and Torralba, Antonio and Tenenbaum, Joshua B},
  journal={arXiv preprint arXiv:2206.01714},
  year={2022}
}