laion-ai/erlich – Run with an API on Replicate

laion-ai / erlich

Generate a logo using text.

Cold

Public
348.9K runs
T4
GitHub
License

Run with an API

Playground API Examples README Versions

Input

prompt

string

Shift + Return to add a new line

paper plane logo with shadow of plane flying around the world, logo, digital artpaper plane logo with shadow of plane flying around the world, logo, digital art

Your text prompt.

Default: ""

negative

string

Shift + Return to add a new line

(optional) Negate the model's prediction for this text from the model's prediction for the target text.

Default: ""

init_image

file

(optional) Initial image to use for the model's prediction. If provided alongside a mask, the image will be inpainted instead.

mask

file

a mask image for inpainting an init_image. white pixels = keep, black pixels = discard. resized to width = image width/8, height = image height/8

guidance_scale

number

(minimum: -20, maximum: 100)

Classifier-free guidance scale. Higher values will result in more guidance toward caption, with diminishing returns. Try values between 1.0 and 40.0. In general, going above 5.0 will introduce some artifacting.

Default: 5

steps

integer

(minimum: 15, maximum: 250)

Number of diffusion steps to run. Due to PLMS sampling, using more than 100 steps is unnecessary and may simply produce the exact same output.

Default: 50

batch_size

integer

(minimum: 1, maximum: 16)

Batch size. (higher = slower)

Default: 4

width

integer

Target width

Default: 256

height

integer

Target height

Default: 256

init_skip_fraction

number

(minimum: 0, maximum: 1)

Fraction of sampling steps to skip when using an init image. Defaults to 0.0 if init_image is not specified and 0.5 if init_image is specified.

Default: 0

aesthetic_rating

integer

Aesthetic rating (1-9) - embed to use.

Default: 9

aesthetic_weight

number

Aesthetic weight (0-1). How much to guide towards the aesthetic embed vs the prompt embed.

Default: 0.5

seed

integer

(minimum: -1, maximum: 4294967295)

Seed for random number generator. If -1, a random seed will be chosen.

Default: -1

intermediate_outputs

boolean

Whether to return intermediate outputs. Enable to visualize the diffusion process and/or debug the model. May slow down inference.

Default: false

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

npx create-replicate --model=laion-ai/erlich

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

const output = await replicate.run(
  "laion-ai/erlich:92fa143ccefeed01534d5d6648bd47796ef06847a6bc55c0e5c5b6975f2dcdfb",
  {
    input: {
      seed: -1,
      steps: 100,
      width: 256,
      height: 256,
      prompt: "paper plane logo with shadow of plane flying around the world, logo, digital art",
      negative: "",
      batch_size: 6,
      guidance_scale: 5,
      aesthetic_rating: 9,
      aesthetic_weight: 0.1,
      init_skip_fraction: 0,
      intermediate_outputs: false
    }
  }
);
console.log(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 laion-ai/erlich using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "laion-ai/erlich:92fa143ccefeed01534d5d6648bd47796ef06847a6bc55c0e5c5b6975f2dcdfb",
    input={
        "seed": -1,
        "steps": 100,
        "width": 256,
        "height": 256,
        "prompt": "paper plane logo with shadow of plane flying around the world, logo, digital art",
        "negative": "",
        "batch_size": 6,
        "guidance_scale": 5,
        "aesthetic_rating": 9,
        "aesthetic_weight": 0.1,
        "init_skip_fraction": 0,
        "intermediate_outputs": False
    }
)

# The laion-ai/erlich model can stream output as it's running.
# The predict method returns an iterator, and you can iterate over that output.
for item in output:
    # https://replicate.com/laion-ai/erlich/api#output-schema
    print(item)

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 laion-ai/erlich 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": "92fa143ccefeed01534d5d6648bd47796ef06847a6bc55c0e5c5b6975f2dcdfb",
    "input": {
      "seed": -1,
      "steps": 100,
      "width": 256,
      "height": 256,
      "prompt": "paper plane logo with shadow of plane flying around the world, logo, digital art",
      "negative": "",
      "batch_size": 6,
      "guidance_scale": 5,
      "aesthetic_rating": 9,
      "aesthetic_weight": 0.1,
      "init_skip_fraction": 0,
      "intermediate_outputs": false
    }
  }' \
  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/laion-ai/erlich@sha256:92fa143ccefeed01534d5d6648bd47796ef06847a6bc55c0e5c5b6975f2dcdfb \
  -i 'seed=-1' \
  -i 'steps=100' \
  -i 'width=256' \
  -i 'height=256' \
  -i 'prompt="paper plane logo with shadow of plane flying around the world, logo, digital art"' \
  -i 'negative=""' \
  -i 'batch_size=6' \
  -i 'guidance_scale=5' \
  -i 'aesthetic_rating=9' \
  -i 'aesthetic_weight=0.1' \
  -i 'init_skip_fraction=0' \
  -i 'intermediate_outputs=false'

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/laion-ai/erlich@sha256:92fa143ccefeed01534d5d6648bd47796ef06847a6bc55c0e5c5b6975f2dcdfb
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "seed": -1,
      "steps": 100,
      "width": 256,
      "height": 256,
      "prompt": "paper plane logo with shadow of plane flying around the world, logo, digital art",
      "negative": "",
      "batch_size": 6,
      "guidance_scale": 5,
      "aesthetic_rating": 9,
      "aesthetic_weight": 0.1,
      "init_skip_fraction": 0,
      "intermediate_outputs": false
    }
  }' \
  http://localhost:5000/predictions

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

Output

{
  "completed_at": "2022-08-05T14:40:21.390360Z",
  "created_at": "2022-08-05T14:33:58.632235Z",
  "data_removed": false,
  "error": null,
  "id": "kjo2hwofv5ak3ffw2673cdnlny",
  "input": {
    "seed": "-1",
    "steps": "100",
    "width": "256",
    "height": "256",
    "prompt": "paper plane logo with shadow of plane flying around the world, logo, digital art",
    "batch_size": "6",
    "guidance_scale": "5",
    "aesthetic_rating": 9,
    "aesthetic_weight": 0.1
  },
  "logs": "Using seed 4158679567\nUsing preloaded models\nEncoding text embeddings with paper plane logo with shadow of plane flying around the world, logo, digital art dimensions\nUsing aesthetic embedding 9 with weight 0.1\nRunning diffusion...\n\n  0%|          | 0/100 [00:00<?, ?it/s]\n  1%|          | 1/100 [00:01<02:28,  1.50s/it]\n  2%|▏         | 2/100 [00:02<02:08,  1.31s/it]\n  3%|▎         | 3/100 [00:03<02:01,  1.25s/it]\n  4%|▍         | 4/100 [00:04<01:23,  1.14it/s]\n  5%|▌         | 5/100 [00:04<01:03,  1.50it/s]\n  6%|▌         | 6/100 [00:04<00:50,  1.86it/s]\n  7%|▋         | 7/100 [00:05<00:42,  2.18it/s]\n  8%|▊         | 8/100 [00:05<00:37,  2.46it/s]\n  9%|▉         | 9/100 [00:05<00:33,  2.69it/s]\n 10%|█         | 10/100 [00:05<00:31,  2.87it/s]\n 11%|█         | 11/100 [00:06<00:29,  3.01it/s]\n 12%|█▏        | 12/100 [00:06<00:28,  3.11it/s]\n 13%|█▎        | 13/100 [00:06<00:27,  3.19it/s]\n 14%|█▍        | 14/100 [00:07<00:26,  3.24it/s]\n 15%|█▌        | 15/100 [00:07<00:25,  3.28it/s]\n 16%|█▌        | 16/100 [00:07<00:25,  3.31it/s]\n 17%|█▋        | 17/100 [00:08<00:24,  3.32it/s]\n 18%|█▊        | 18/100 [00:08<00:24,  3.34it/s]\n 19%|█▉        | 19/100 [00:08<00:24,  3.35it/s]\n 20%|██        | 20/100 [00:08<00:23,  3.35it/s]\n 21%|██        | 21/100 [00:09<00:23,  3.36it/s]\n 22%|██▏       | 22/100 [00:09<00:23,  3.37it/s]\n 23%|██▎       | 23/100 [00:09<00:22,  3.39it/s]\n 24%|██▍       | 24/100 [00:10<00:22,  3.38it/s]\n 25%|██▌       | 25/100 [00:10<00:22,  3.38it/s]\n 26%|██▌       | 26/100 [00:10<00:21,  3.37it/s]\n 27%|██▋       | 27/100 [00:10<00:21,  3.36it/s]\n 28%|██▊       | 28/100 [00:11<00:21,  3.36it/s]\n 29%|██▉       | 29/100 [00:11<00:21,  3.36it/s]\n 30%|███       | 30/100 [00:11<00:20,  3.37it/s]\n 31%|███       | 31/100 [00:12<00:20,  3.37it/s]\n 32%|███▏      | 32/100 [00:12<00:20,  3.37it/s]\n 33%|███▎      | 33/100 [00:12<00:19,  3.38it/s]\n 34%|███▍      | 34/100 [00:13<00:19,  3.37it/s]\n 35%|███▌      | 35/100 [00:13<00:19,  3.38it/s]\n 36%|███▌      | 36/100 [00:13<00:18,  3.38it/s]\n 37%|███▋      | 37/100 [00:13<00:18,  3.37it/s]\n 38%|███▊      | 38/100 [00:14<00:18,  3.37it/s]\n 39%|███▉      | 39/100 [00:14<00:18,  3.37it/s]\n 40%|████      | 40/100 [00:14<00:17,  3.37it/s]\n 41%|████      | 41/100 [00:15<00:17,  3.37it/s]\n 42%|████▏     | 42/100 [00:15<00:17,  3.37it/s]\n 43%|████▎     | 43/100 [00:15<00:16,  3.36it/s]\n 44%|████▍     | 44/100 [00:16<00:16,  3.36it/s]\n 45%|████▌     | 45/100 [00:16<00:16,  3.36it/s]\n 46%|████▌     | 46/100 [00:16<00:16,  3.36it/s]\n 47%|████▋     | 47/100 [00:16<00:15,  3.36it/s]\n 48%|████▊     | 48/100 [00:17<00:15,  3.35it/s]\n 49%|████▉     | 49/100 [00:17<00:15,  3.36it/s]\n 50%|█████     | 50/100 [00:17<00:14,  3.35it/s]\n 51%|█████     | 51/100 [00:18<00:14,  3.35it/s]\n 52%|█████▏    | 52/100 [00:18<00:14,  3.34it/s]\n 53%|█████▎    | 53/100 [00:18<00:14,  3.35it/s]\n 54%|█████▍    | 54/100 [00:18<00:13,  3.35it/s]\n 55%|█████▌    | 55/100 [00:19<00:13,  3.34it/s]\n 56%|█████▌    | 56/100 [00:19<00:13,  3.35it/s]\n 57%|█████▋    | 57/100 [00:19<00:12,  3.35it/s]\n 58%|█████▊    | 58/100 [00:20<00:12,  3.35it/s]\n 59%|█████▉    | 59/100 [00:20<00:12,  3.35it/s]\n 60%|██████    | 60/100 [00:20<00:11,  3.34it/s]\n 61%|██████    | 61/100 [00:21<00:11,  3.34it/s]\n 62%|██████▏   | 62/100 [00:21<00:11,  3.34it/s]\n 63%|██████▎   | 63/100 [00:21<00:11,  3.33it/s]\n 64%|██████▍   | 64/100 [00:21<00:10,  3.34it/s]\n 65%|██████▌   | 65/100 [00:22<00:10,  3.33it/s]\n 66%|██████▌   | 66/100 [00:22<00:10,  3.33it/s]\n 67%|██████▋   | 67/100 [00:22<00:09,  3.34it/s]\n 68%|██████▊   | 68/100 [00:23<00:09,  3.34it/s]\n 69%|██████▉   | 69/100 [00:23<00:09,  3.34it/s]\n 70%|███████   | 70/100 [00:23<00:09,  3.33it/s]\n 71%|███████   | 71/100 [00:24<00:08,  3.32it/s]\n 72%|███████▏  | 72/100 [00:24<00:08,  3.32it/s]\n 73%|███████▎  | 73/100 [00:24<00:08,  3.32it/s]\n 74%|███████▍  | 74/100 [00:24<00:07,  3.32it/s]\n 75%|███████▌  | 75/100 [00:25<00:07,  3.33it/s]\n 76%|███████▌  | 76/100 [00:25<00:07,  3.33it/s]\n 77%|███████▋  | 77/100 [00:25<00:06,  3.34it/s]\n 78%|███████▊  | 78/100 [00:26<00:06,  3.33it/s]\n 79%|███████▉  | 79/100 [00:26<00:06,  3.34it/s]\n 80%|████████  | 80/100 [00:26<00:05,  3.33it/s]\n 81%|████████  | 81/100 [00:27<00:05,  3.33it/s]\n 82%|████████▏ | 82/100 [00:27<00:05,  3.33it/s]\n 83%|████████▎ | 83/100 [00:27<00:05,  3.33it/s]\n 84%|████████▍ | 84/100 [00:27<00:04,  3.33it/s]\n 85%|████████▌ | 85/100 [00:28<00:04,  3.33it/s]\n 86%|████████▌ | 86/100 [00:28<00:04,  3.32it/s]\n 87%|████████▋ | 87/100 [00:28<00:03,  3.32it/s]\n 88%|████████▊ | 88/100 [00:29<00:03,  3.32it/s]\n 89%|████████▉ | 89/100 [00:29<00:03,  3.32it/s]\n 90%|█████████ | 90/100 [00:29<00:03,  3.32it/s]\n 91%|█████████ | 91/100 [00:30<00:02,  3.32it/s]\n 92%|█████████▏| 92/100 [00:30<00:02,  3.32it/s]\n 93%|█████████▎| 93/100 [00:30<00:02,  3.32it/s]\n 94%|█████████▍| 94/100 [00:31<00:01,  3.32it/s]\n 95%|█████████▌| 95/100 [00:31<00:01,  3.32it/s]\n 96%|█████████▌| 96/100 [00:31<00:01,  3.33it/s]\n 97%|█████████▋| 97/100 [00:31<00:00,  3.32it/s]\n 98%|█████████▊| 98/100 [00:32<00:00,  3.33it/s]\n 99%|█████████▉| 99/100 [00:32<00:00,  3.32it/s]\n100%|██████████| 100/100 [00:32<00:00,  3.32it/s]\n100%|██████████| 100/100 [00:32<00:00,  3.05it/s]\nSaving final sample/s",
  "metrics": {
    "predict_time": 37.033327,
    "total_time": 382.758125
  },
  "output": [
    [
      "https://replicate.delivery/mgxm/f71aaee1-8937-403c-b497-9d05921cba95/ts_99-batch_0.png",
      "https://replicate.delivery/mgxm/0dae8f90-d30f-4bc6-a21d-98d2e6918ba5/ts_99-batch_1.png",
      "https://replicate.delivery/mgxm/3d7569c9-e1d1-400e-a7ad-1ab97c98c97c/ts_99-batch_2.png",
      "https://replicate.delivery/mgxm/264565dc-6828-4fa1-8521-614ff5c236fa/ts_99-batch_3.png",
      "https://replicate.delivery/mgxm/8022084b-a249-4979-ab0f-c774f687cf77/ts_99-batch_4.png",
      "https://replicate.delivery/mgxm/bcfb9294-1da5-4762-b761-dd14ad601f03/ts_99-batch_5.png"
    ]
  ],
  "started_at": "2022-08-05T14:39:44.357033Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/kjo2hwofv5ak3ffw2673cdnlny",
    "cancel": "https://api.replicate.com/v1/predictions/kjo2hwofv5ak3ffw2673cdnlny/cancel"
  },
  "version": "92fa143ccefeed01534d5d6648bd47796ef06847a6bc55c0e5c5b6975f2dcdfb"
}

Generated in

37.0 seconds

Tweak itReport

Using seed 4158679567
Using preloaded models
Encoding text embeddings with paper plane logo with shadow of plane flying around the world, logo, digital art dimensions
Using aesthetic embedding 9 with weight 0.1
Running diffusion...

  0%|          | 0/100 [00:00<?, ?it/s]
  1%|          | 1/100 [00:01<02:28,  1.50s/it]
  2%|▏         | 2/100 [00:02<02:08,  1.31s/it]
  3%|▎         | 3/100 [00:03<02:01,  1.25s/it]
  4%|▍         | 4/100 [00:04<01:23,  1.14it/s]
  5%|▌         | 5/100 [00:04<01:03,  1.50it/s]
  6%|▌         | 6/100 [00:04<00:50,  1.86it/s]
  7%|▋         | 7/100 [00:05<00:42,  2.18it/s]
  8%|▊         | 8/100 [00:05<00:37,  2.46it/s]
  9%|▉         | 9/100 [00:05<00:33,  2.69it/s]
 10%|█         | 10/100 [00:05<00:31,  2.87it/s]
 11%|█         | 11/100 [00:06<00:29,  3.01it/s]
 12%|█▏        | 12/100 [00:06<00:28,  3.11it/s]
 13%|█▎        | 13/100 [00:06<00:27,  3.19it/s]
 14%|█▍        | 14/100 [00:07<00:26,  3.24it/s]
 15%|█▌        | 15/100 [00:07<00:25,  3.28it/s]
 16%|█▌        | 16/100 [00:07<00:25,  3.31it/s]
 17%|█▋        | 17/100 [00:08<00:24,  3.32it/s]
 18%|█▊        | 18/100 [00:08<00:24,  3.34it/s]
 19%|█▉        | 19/100 [00:08<00:24,  3.35it/s]
 20%|██        | 20/100 [00:08<00:23,  3.35it/s]
 21%|██        | 21/100 [00:09<00:23,  3.36it/s]
 22%|██▏       | 22/100 [00:09<00:23,  3.37it/s]
 23%|██▎       | 23/100 [00:09<00:22,  3.39it/s]
 24%|██▍       | 24/100 [00:10<00:22,  3.38it/s]
 25%|██▌       | 25/100 [00:10<00:22,  3.38it/s]
 26%|██▌       | 26/100 [00:10<00:21,  3.37it/s]
 27%|██▋       | 27/100 [00:10<00:21,  3.36it/s]
 28%|██▊       | 28/100 [00:11<00:21,  3.36it/s]
 29%|██▉       | 29/100 [00:11<00:21,  3.36it/s]
 30%|███       | 30/100 [00:11<00:20,  3.37it/s]
 31%|███       | 31/100 [00:12<00:20,  3.37it/s]
 32%|███▏      | 32/100 [00:12<00:20,  3.37it/s]
 33%|███▎      | 33/100 [00:12<00:19,  3.38it/s]
 34%|███▍      | 34/100 [00:13<00:19,  3.37it/s]
 35%|███▌      | 35/100 [00:13<00:19,  3.38it/s]
 36%|███▌      | 36/100 [00:13<00:18,  3.38it/s]
 37%|███▋      | 37/100 [00:13<00:18,  3.37it/s]
 38%|███▊      | 38/100 [00:14<00:18,  3.37it/s]
 39%|███▉      | 39/100 [00:14<00:18,  3.37it/s]
 40%|████      | 40/100 [00:14<00:17,  3.37it/s]
 41%|████      | 41/100 [00:15<00:17,  3.37it/s]
 42%|████▏     | 42/100 [00:15<00:17,  3.37it/s]
 43%|████▎     | 43/100 [00:15<00:16,  3.36it/s]
 44%|████▍     | 44/100 [00:16<00:16,  3.36it/s]
 45%|████▌     | 45/100 [00:16<00:16,  3.36it/s]
 46%|████▌     | 46/100 [00:16<00:16,  3.36it/s]
 47%|████▋     | 47/100 [00:16<00:15,  3.36it/s]
 48%|████▊     | 48/100 [00:17<00:15,  3.35it/s]
 49%|████▉     | 49/100 [00:17<00:15,  3.36it/s]
 50%|█████     | 50/100 [00:17<00:14,  3.35it/s]
 51%|█████     | 51/100 [00:18<00:14,  3.35it/s]
 52%|█████▏    | 52/100 [00:18<00:14,  3.34it/s]
 53%|█████▎    | 53/100 [00:18<00:14,  3.35it/s]
 54%|█████▍    | 54/100 [00:18<00:13,  3.35it/s]
 55%|█████▌    | 55/100 [00:19<00:13,  3.34it/s]
 56%|█████▌    | 56/100 [00:19<00:13,  3.35it/s]
 57%|█████▋    | 57/100 [00:19<00:12,  3.35it/s]
 58%|█████▊    | 58/100 [00:20<00:12,  3.35it/s]
 59%|█████▉    | 59/100 [00:20<00:12,  3.35it/s]
 60%|██████    | 60/100 [00:20<00:11,  3.34it/s]
 61%|██████    | 61/100 [00:21<00:11,  3.34it/s]
 62%|██████▏   | 62/100 [00:21<00:11,  3.34it/s]
 63%|██████▎   | 63/100 [00:21<00:11,  3.33it/s]
 64%|██████▍   | 64/100 [00:21<00:10,  3.34it/s]
 65%|██████▌   | 65/100 [00:22<00:10,  3.33it/s]
 66%|██████▌   | 66/100 [00:22<00:10,  3.33it/s]
 67%|██████▋   | 67/100 [00:22<00:09,  3.34it/s]
 68%|██████▊   | 68/100 [00:23<00:09,  3.34it/s]
 69%|██████▉   | 69/100 [00:23<00:09,  3.34it/s]
 70%|███████   | 70/100 [00:23<00:09,  3.33it/s]
 71%|███████   | 71/100 [00:24<00:08,  3.32it/s]
 72%|███████▏  | 72/100 [00:24<00:08,  3.32it/s]
 73%|███████▎  | 73/100 [00:24<00:08,  3.32it/s]
 74%|███████▍  | 74/100 [00:24<00:07,  3.32it/s]
 75%|███████▌  | 75/100 [00:25<00:07,  3.33it/s]
 76%|███████▌  | 76/100 [00:25<00:07,  3.33it/s]
 77%|███████▋  | 77/100 [00:25<00:06,  3.34it/s]
 78%|███████▊  | 78/100 [00:26<00:06,  3.33it/s]
 79%|███████▉  | 79/100 [00:26<00:06,  3.34it/s]
 80%|████████  | 80/100 [00:26<00:05,  3.33it/s]
 81%|████████  | 81/100 [00:27<00:05,  3.33it/s]
 82%|████████▏ | 82/100 [00:27<00:05,  3.33it/s]
 83%|████████▎ | 83/100 [00:27<00:05,  3.33it/s]
 84%|████████▍ | 84/100 [00:27<00:04,  3.33it/s]
 85%|████████▌ | 85/100 [00:28<00:04,  3.33it/s]
 86%|████████▌ | 86/100 [00:28<00:04,  3.32it/s]
 87%|████████▋ | 87/100 [00:28<00:03,  3.32it/s]
 88%|████████▊ | 88/100 [00:29<00:03,  3.32it/s]
 89%|████████▉ | 89/100 [00:29<00:03,  3.32it/s]
 90%|█████████ | 90/100 [00:29<00:03,  3.32it/s]
 91%|█████████ | 91/100 [00:30<00:02,  3.32it/s]
 92%|█████████▏| 92/100 [00:30<00:02,  3.32it/s]
 93%|█████████▎| 93/100 [00:30<00:02,  3.32it/s]
 94%|█████████▍| 94/100 [00:31<00:01,  3.32it/s]
 95%|█████████▌| 95/100 [00:31<00:01,  3.32it/s]
 96%|█████████▌| 96/100 [00:31<00:01,  3.33it/s]
 97%|█████████▋| 97/100 [00:31<00:00,  3.32it/s]
 98%|█████████▊| 98/100 [00:32<00:00,  3.33it/s]
 99%|█████████▉| 99/100 [00:32<00:00,  3.32it/s]
100%|██████████| 100/100 [00:32<00:00,  3.32it/s]
100%|██████████| 100/100 [00:32<00:00,  3.05it/s]
Saving final sample/s

Examples

View more examples

Run time and cost

This model costs approximately $0.062 to run on Replicate, or 16 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 5 minutes. The predict time for this model varies significantly based on the inputs.

Readme

erlich is the text2image latent diffusion model from CompVis (with additions from glid-3-xl) finetuned on a dataset collected from LAION-5B named Large Logo Dataset. It consists of roughly 100K images of logos with captions generated via BLIP using aggressive re-ranking.

For more info see the README.md for ldm-finetune.