nicholascelestin/dalle-mega – Run with an API on Replicate

nicholascelestin / dalle-mega

Made public only for API calls. Use min-dalle instead-- it's superior.

Cold

Public
12.1K runs
T4
GitHub
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=nicholascelestin/dalle-mega

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 nicholascelestin/dalle-mega using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "nicholascelestin/dalle-mega:b5880117d39b1cd84efae7ceb201231463b93b086843d73273db7ee7b1decdd2",
  {
    input: {
      num: 10,
      prompt: "A fantasy kingdom",
      model_size: "MINI"
    }
  }
);

// 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 nicholascelestin/dalle-mega using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "nicholascelestin/dalle-mega:b5880117d39b1cd84efae7ceb201231463b93b086843d73273db7ee7b1decdd2",
    input={
        "num": 10,
        "prompt": "A fantasy kingdom",
        "model_size": "MINI"
    }
)
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 nicholascelestin/dalle-mega 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": "nicholascelestin/dalle-mega:b5880117d39b1cd84efae7ceb201231463b93b086843d73273db7ee7b1decdd2",
    "input": {
      "num": 10,
      "prompt": "A fantasy kingdom",
      "model_size": "MINI"
    }
  }' \
  https://api.replicate.com/v1/predictions

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

Output

{
  "completed_at": "2022-06-15T21:32:57.476643Z",
  "created_at": "2022-06-15T21:28:59.725063Z",
  "data_removed": false,
  "error": null,
  "id": "nou7sr6jzvd75gtx3uo524txcm",
  "input": {
    "num": 10,
    "prompt": "A fantasy kingdom",
    "model_size": "MINI"
  },
  "logs": "Wed Jun 15 21:31:59 2022\n+-----------------------------------------------------------------------------+\n| NVIDIA-SMI 450.119.04   Driver Version: 450.119.04   CUDA Version: 11.0     |\n|-------------------------------+----------------------+----------------------+\n| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |\n| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |\n|                               |                      |               MIG M. |\n|===============================+======================+======================|\n|   0  Tesla T4            Off  | 00000000:00:04.0 Off |                    0 |\n| N/A   64C    P0    30W /  70W |   2140MiB / 15109MiB |     14%      Default |\n|                               |                      |                  N/A |\n+-------------------------------+----------------------+----------------------+\n\n+-----------------------------------------------------------------------------+\n| Processes:                                                                  |\n|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |\n|        ID   ID                                                   Usage      |\n|=============================================================================|\n+-----------------------------------------------------------------------------+\n\nTokenizing at 4.76837158203125e-06\nReplicating at 0.06999850273132324\nGenerating images at 0.07134580612182617\nencoding image 0\nencoded image 0\ndecoding image 0\ndecoded image 0\nsaving image 0\nimage 0 saved to 2aad91ed-9894-4534-9197-7dfbbef4538c.png\ntook 28.273508548736572\nencoding image 1\nencoded image 1\ndecoding image 1\ndecoded image 1\nsaving image 1\nimage 1 saved to 1113663e-e2ac-4cdf-8848-77afee1d2255.png\ntook 31.464338064193726\nencoding image 2\nencoded image 2\ndecoding image 2\ndecoded image 2\nsaving image 2\nimage 2 saved to f78f4d82-c800-4a86-9df0-f1947096fec5.png\ntook 34.615535497665405\nencoding image 3\nencoded image 3\ndecoding image 3\ndecoded image 3\nsaving image 3\nimage 3 saved to 32016a67-35cc-4b5d-9067-482c212fc89e.png\ntook 37.762876749038696\nencoding image 4\nencoded image 4\ndecoding image 4\ndecoded image 4\nsaving image 4\nimage 4 saved to 0fa6ef63-cf67-4ec1-a396-6f8834e06906.png\ntook 40.92224478721619\nencoding image 5\nencoded image 5\ndecoding image 5\ndecoded image 5\nsaving image 5\nimage 5 saved to 851d611c-b2fa-47c2-9164-f683025db2a5.png\ntook 44.092308044433594\nencoding image 6\nencoded image 6\ndecoding image 6\ndecoded image 6\nsaving image 6\nimage 6 saved to bfb1bc4d-a740-4bbe-ad9e-c41b74e037fc.png\ntook 47.26582431793213\nencoding image 7\nencoded image 7\ndecoding image 7\ndecoded image 7\nsaving image 7\nimage 7 saved to b41acc06-052a-420e-89f6-14df6a82d308.png\ntook 50.447410345077515\nencoding image 8\nencoded image 8\ndecoding image 8\ndecoded image 8\nsaving image 8\nimage 8 saved to 3c5f91ad-5a5c-4613-93fc-eda58dcff934.png\ntook 53.6339430809021\nencoding image 9\nencoded image 9\ndecoding image 9\ndecoded image 9\nsaving image 9\nimage 9 saved to b99bcb06-326a-4262-8070-9980429bd3ad.png\ntook 56.82738733291626",
  "metrics": {
    "predict_time": 56.886768,
    "total_time": 237.75158
  },
  "output": [
    "https://replicate.delivery/mgxm/117d2d2d-8912-4380-8d24-fac01a85bfd8/2aad91ed-9894-4534-9197-7dfbbef4538c.png",
    "https://replicate.delivery/mgxm/c3f94167-3785-4a00-a9aa-762d1b3efa5b/1113663e-e2ac-4cdf-8848-77afee1d2255.png",
    "https://replicate.delivery/mgxm/084b960d-e4e1-448d-95c2-db1f135a1e0b/f78f4d82-c800-4a86-9df0-f1947096fec5.png",
    "https://replicate.delivery/mgxm/9417d9ae-1969-4a8c-95e6-4522a526ddd3/32016a67-35cc-4b5d-9067-482c212fc89e.png",
    "https://replicate.delivery/mgxm/09a937b7-5449-4e01-a894-efb47a897cb8/0fa6ef63-cf67-4ec1-a396-6f8834e06906.png",
    "https://replicate.delivery/mgxm/c15bcbb9-5d8b-4be6-aaf6-b190997ce5f1/851d611c-b2fa-47c2-9164-f683025db2a5.png",
    "https://replicate.delivery/mgxm/e53e94f9-e17c-407a-badb-1932bb78235d/bfb1bc4d-a740-4bbe-ad9e-c41b74e037fc.png",
    "https://replicate.delivery/mgxm/aabdd8e2-463b-47de-870d-c78eb6d26387/b41acc06-052a-420e-89f6-14df6a82d308.png",
    "https://replicate.delivery/mgxm/d5b48e04-544b-4a5b-9708-611df3cb67d2/3c5f91ad-5a5c-4613-93fc-eda58dcff934.png",
    "https://replicate.delivery/mgxm/4c5b72c5-1ad7-4470-bd20-32f0a38c2299/b99bcb06-326a-4262-8070-9980429bd3ad.png"
  ],
  "started_at": "2022-06-15T21:32:00.589875Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/nou7sr6jzvd75gtx3uo524txcm",
    "cancel": "https://api.replicate.com/v1/predictions/nou7sr6jzvd75gtx3uo524txcm/cancel"
  },
  "version": "70dbc2347bdc08d78af32a0fca8c1cafd3a6a7c29ea2bfa5a2fbbf0e70a1e1b8"
}

Generated in

56.9 seconds

Tweak itReport

Wed Jun 15 21:31:59 2022
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 450.119.04   Driver Version: 450.119.04   CUDA Version: 11.0     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  Tesla T4            Off  | 00000000:00:04.0 Off |                    0 |
| N/A   64C    P0    30W /  70W |   2140MiB / 15109MiB |     14%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
+-----------------------------------------------------------------------------+

Tokenizing at 4.76837158203125e-06
Replicating at 0.06999850273132324
Generating images at 0.07134580612182617
encoding image 0
encoded image 0
decoding image 0
decoded image 0
saving image 0
image 0 saved to 2aad91ed-9894-4534-9197-7dfbbef4538c.png
took 28.273508548736572
encoding image 1
encoded image 1
decoding image 1
decoded image 1
saving image 1
image 1 saved to 1113663e-e2ac-4cdf-8848-77afee1d2255.png
took 31.464338064193726
encoding image 2
encoded image 2
decoding image 2
decoded image 2
saving image 2
image 2 saved to f78f4d82-c800-4a86-9df0-f1947096fec5.png
took 34.615535497665405
encoding image 3
encoded image 3
decoding image 3
decoded image 3
saving image 3
image 3 saved to 32016a67-35cc-4b5d-9067-482c212fc89e.png
took 37.762876749038696
encoding image 4
encoded image 4
decoding image 4
decoded image 4
saving image 4
image 4 saved to 0fa6ef63-cf67-4ec1-a396-6f8834e06906.png
took 40.92224478721619
encoding image 5
encoded image 5
decoding image 5
decoded image 5
saving image 5
image 5 saved to 851d611c-b2fa-47c2-9164-f683025db2a5.png
took 44.092308044433594
encoding image 6
encoded image 6
decoding image 6
decoded image 6
saving image 6
image 6 saved to bfb1bc4d-a740-4bbe-ad9e-c41b74e037fc.png
took 47.26582431793213
encoding image 7
encoded image 7
decoding image 7
decoded image 7
saving image 7
image 7 saved to b41acc06-052a-420e-89f6-14df6a82d308.png
took 50.447410345077515
encoding image 8
encoded image 8
decoding image 8
decoded image 8
saving image 8
image 8 saved to 3c5f91ad-5a5c-4613-93fc-eda58dcff934.png
took 53.6339430809021
encoding image 9
encoded image 9
decoding image 9
decoded image 9
saving image 9
image 9 saved to b99bcb06-326a-4262-8070-9980429bd3ad.png
took 56.82738733291626

This output was created using a different version of the model, nicholascelestin/dalle-mega:70dbc234.

Examples

View more examples

Run time and cost

This model runs on Nvidia T4 GPU hardware. We don't yet have enough runs of this model to provide performance information.

Readme

This is a quick port of DALLE-mini/mega to replicate, and the official repository is here https://github.com/borisdayma/dalle-mini . This model is much quicker than GLID or PixRay– good for generating a dozen images at once. But lower quality.