aisha-ai-official/anillustrious-v4 | Run with an API on Replicate

aisha-ai-official / anillustrious-v4

Warm

Public
145.3K runs
L40S

Iterate in playground

Run with an API

Playground API Examples README Versions

Input

model

string

The model to use

Default: "Anillustrious-v4"

vae

string

The VAE to use

Default: "default"

prompt_conjunction

boolean

Force Compel conjunction syntax on prompt, separating linebreaks into different items on the Compel conjunction function

Default: false

prompt

string

Shift + Return to add a new line

from above,street
1girl, dark-purple short hair, purple eyes, medium breasts, cleavage, casual clothes, smile
from above,street
1girl, dark-purple short hair, purple eyes, medium breasts, cleavage, casual clothes, smile

The prompt, uses Compel weighting syntax

Default: "street, 1girl, dark-purple short hair, purple eyes, medium breasts, cleavage, casual clothes, smile, V"

negative_prompt

string

Shift + Return to add a new line

The negative prompt (for things you don't want), uses Compel weighting syntax

Default: "nsfw, naked"

negative_prompt_conjunction

boolean

Force Compel conjunction syntax on prompt, separating linebreaks into different items on the Compel conjunction function

Default: false

cfg_scale

number

(minimum: 1, maximum: 50)

CFG scale defines how much attention the model pays to the prompt when generating, set to 1 to disable

Default: 7

guidance_rescale

number

(minimum: 0, maximum: 5)

Amount to rescale CFG generated noise to avoid overexposed images, set to 0 or 1 to disable

Default: 1

pag_scale

number

(minimum: 0, maximum: 50)

PAG scale is similar to CFG but improves the result; set to 0 to disable

Default: 0

clip_skip

integer

(minimum: 0)

How many CLIP layers to skip; 1 means no skip

Default: 2

width

integer

(minimum: 1, maximum: 4096)

The width of the image

Default: 1024

height

integer

(minimum: 1, maximum: 4096)

The height of the image

Default: 1024

upscale

string

Upscale factor

Default: "Original"

refiner

boolean

Enable image refiner using model

Default: false

refiner_prompt

string

Shift + Return to add a new line

The prompt of refiner (leave empty for default correct hands and details)

Default: ""

refiner_strength

number

(minimum: 0, maximum: 1)

Refiner strength (0.0 to 1.0)

Default: 0.4

prepend_preprompt

boolean

Prepend preprompt (Prompt: "masterpiece, best_quality, amazing_quality, very_aesthetic, absurdres, newest," | Negative prompt: "lowres, worst quality, low quality, bad anatomy, worst aesthetic, jpeg artifacts, scan artifacts, compression artifacts,")

Default: true

scheduler

string

The scheduler to use

Default: "LCMScheduler Karras"

steps

integer

(minimum: 1, maximum: 100)

Number of steps when generating

Default: 16

seed

integer

The seed for generation; set to -1 for random seed

Default: -1

adetailer_face

boolean

Use ADetailer face_yolov9c.pt

Default: false

adetailer_face_prompt

string

Shift + Return to add a new line

ADetailer face prompt (if blank, will use the image prompt)

Default: ""

adetailer_face_negative_prompt

string

Shift + Return to add a new line

ADetailer face negative prompt (if blank, will use: blurry, deformed face, extra features, unrealistic skin texture, oversmoothed, low resolution, artifacts)

Default: ""

adetailer_hand

boolean

Use ADetailer hand_yolov9c.pt

Default: false

adetailer_hand_prompt

string

Shift + Return to add a new line

ADetailer hand prompt (if blank, will use: detailed hands, realistic anatomy, well-defined fingers, natural pose, soft shadows, high resolution, intricate details)

Default: ""

adetailer_hand_negative_prompt

string

Shift + Return to add a new line

ADetailer hand negative prompt (if blank, will use: extra fingers, fused or missing fingers, blurry, distorted, low detail, poorly drawn, unnatural proportions)

Default: ""

adetailer_person

boolean

Use ADetailer person_yolov8m-seg.pt

Default: false

adetailer_person_prompt

string

Shift + Return to add a new line

ADetailer person prompt (if blank, will use: best quality, detailed, realistic shadows)

Default: ""

adetailer_person_negative_prompt

string

Shift + Return to add a new line

ADetailer person negative prompt (if blank, will use: deformed, unrealistic proportions, extra limbs, blurred, low quality, distorted pose, exaggerated features)

Default: ""

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

npx create-replicate --model=aisha-ai-official/anillustrious-v4

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 aisha-ai-official/anillustrious-v4 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "aisha-ai-official/anillustrious-v4:80441e2c32a55f2fcf9b77fa0a74c6c86ad7deac51eed722b9faedb253265cb4",
  {
    input: {
      vae: "default",
      seed: -1,
      model: "Anillustrious-v4",
      steps: 30,
      width: 1024,
      height: 1024,
      prompt: "\nfrom above,street\n1girl, dark-purple short hair, purple eyes, medium breasts, cleavage, casual clothes, smile",
      refiner: false,
      upscale: "Original",
      cfg_scale: 7,
      clip_skip: 2,
      pag_scale: 0,
      scheduler: "Euler a beta",
      adetailer_face: false,
      adetailer_hand: false,
      refiner_prompt: "",
      negative_prompt: "nsfw, naked",
      adetailer_person: false,
      guidance_rescale: 1,
      refiner_strength: 0.8,
      prepend_preprompt: true,
      prompt_conjunction: true,
      adetailer_face_prompt: "",
      adetailer_hand_prompt: "",
      adetailer_person_prompt: "",
      negative_prompt_conjunction: false,
      adetailer_face_negative_prompt: "",
      adetailer_hand_negative_prompt: "",
      adetailer_person_negative_prompt: ""
    }
  }
);

// 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 aisha-ai-official/anillustrious-v4 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "aisha-ai-official/anillustrious-v4:80441e2c32a55f2fcf9b77fa0a74c6c86ad7deac51eed722b9faedb253265cb4",
    input={
        "vae": "default",
        "seed": -1,
        "model": "Anillustrious-v4",
        "steps": 30,
        "width": 1024,
        "height": 1024,
        "prompt": "\nfrom above,street\n1girl, dark-purple short hair, purple eyes, medium breasts, cleavage, casual clothes, smile",
        "refiner": False,
        "upscale": "Original",
        "cfg_scale": 7,
        "clip_skip": 2,
        "pag_scale": 0,
        "scheduler": "Euler a beta",
        "adetailer_face": False,
        "adetailer_hand": False,
        "refiner_prompt": "",
        "negative_prompt": "nsfw, naked",
        "adetailer_person": False,
        "guidance_rescale": 1,
        "refiner_strength": 0.8,
        "prepend_preprompt": True,
        "prompt_conjunction": True,
        "adetailer_face_prompt": "",
        "adetailer_hand_prompt": "",
        "adetailer_person_prompt": "",
        "negative_prompt_conjunction": False,
        "adetailer_face_negative_prompt": "",
        "adetailer_hand_negative_prompt": "",
        "adetailer_person_negative_prompt": ""
    }
)
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 aisha-ai-official/anillustrious-v4 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": "aisha-ai-official/anillustrious-v4:80441e2c32a55f2fcf9b77fa0a74c6c86ad7deac51eed722b9faedb253265cb4",
    "input": {
      "vae": "default",
      "seed": -1,
      "model": "Anillustrious-v4",
      "steps": 30,
      "width": 1024,
      "height": 1024,
      "prompt": "\\nfrom above,street\\n1girl, dark-purple short hair, purple eyes, medium breasts, cleavage, casual clothes, smile",
      "refiner": false,
      "upscale": "Original",
      "cfg_scale": 7,
      "clip_skip": 2,
      "pag_scale": 0,
      "scheduler": "Euler a beta",
      "adetailer_face": false,
      "adetailer_hand": false,
      "refiner_prompt": "",
      "negative_prompt": "nsfw, naked",
      "adetailer_person": false,
      "guidance_rescale": 1,
      "refiner_strength": 0.8,
      "prepend_preprompt": true,
      "prompt_conjunction": true,
      "adetailer_face_prompt": "",
      "adetailer_hand_prompt": "",
      "adetailer_person_prompt": "",
      "negative_prompt_conjunction": false,
      "adetailer_face_negative_prompt": "",
      "adetailer_hand_negative_prompt": "",
      "adetailer_person_negative_prompt": ""
    }
  }' \
  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/aisha-ai-official/anillustrious-v4@sha256:80441e2c32a55f2fcf9b77fa0a74c6c86ad7deac51eed722b9faedb253265cb4 \
  -i 'vae="default"' \
  -i 'seed=-1' \
  -i 'model="Anillustrious-v4"' \
  -i 'steps=30' \
  -i 'width=1024' \
  -i 'height=1024' \
  -i $'prompt="\\nfrom above,street\\n1girl, dark-purple short hair, purple eyes, medium breasts, cleavage, casual clothes, smile"' \
  -i 'refiner=false' \
  -i 'upscale="Original"' \
  -i 'cfg_scale=7' \
  -i 'clip_skip=2' \
  -i 'pag_scale=0' \
  -i 'scheduler="Euler a beta"' \
  -i 'adetailer_face=false' \
  -i 'adetailer_hand=false' \
  -i 'refiner_prompt=""' \
  -i 'negative_prompt="nsfw, naked"' \
  -i 'adetailer_person=false' \
  -i 'guidance_rescale=1' \
  -i 'refiner_strength=0.8' \
  -i 'prepend_preprompt=true' \
  -i 'prompt_conjunction=true' \
  -i 'adetailer_face_prompt=""' \
  -i 'adetailer_hand_prompt=""' \
  -i 'adetailer_person_prompt=""' \
  -i 'negative_prompt_conjunction=false' \
  -i 'adetailer_face_negative_prompt=""' \
  -i 'adetailer_hand_negative_prompt=""' \
  -i 'adetailer_person_negative_prompt=""'

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/aisha-ai-official/anillustrious-v4@sha256:80441e2c32a55f2fcf9b77fa0a74c6c86ad7deac51eed722b9faedb253265cb4
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "vae": "default",
      "seed": -1,
      "model": "Anillustrious-v4",
      "steps": 30,
      "width": 1024,
      "height": 1024,
      "prompt": "\\nfrom above,street\\n1girl, dark-purple short hair, purple eyes, medium breasts, cleavage, casual clothes, smile",
      "refiner": false,
      "upscale": "Original",
      "cfg_scale": 7,
      "clip_skip": 2,
      "pag_scale": 0,
      "scheduler": "Euler a beta",
      "adetailer_face": false,
      "adetailer_hand": false,
      "refiner_prompt": "",
      "negative_prompt": "nsfw, naked",
      "adetailer_person": false,
      "guidance_rescale": 1,
      "refiner_strength": 0.8,
      "prepend_preprompt": true,
      "prompt_conjunction": true,
      "adetailer_face_prompt": "",
      "adetailer_hand_prompt": "",
      "adetailer_person_prompt": "",
      "negative_prompt_conjunction": false,
      "adetailer_face_negative_prompt": "",
      "adetailer_hand_negative_prompt": "",
      "adetailer_person_negative_prompt": ""
    }
  }' \
  http://localhost:5000/predictions

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

Output

{
  "completed_at": "2025-03-20T16:10:52.121409Z",
  "created_at": "2025-03-20T16:10:45.905000Z",
  "data_removed": false,
  "error": null,
  "id": "fw5b2a2725rm80cnph9b96dp44",
  "input": {
    "vae": "default",
    "seed": -1,
    "model": "Anillustrious-v4",
    "steps": 30,
    "width": 1024,
    "height": 1024,
    "prompt": "\nfrom above,street\n1girl, dark-purple short hair, purple eyes, medium breasts, cleavage, casual clothes, smile",
    "refiner": false,
    "upscale": "Original",
    "cfg_scale": 7,
    "clip_skip": 2,
    "pag_scale": 0,
    "scheduler": "Euler a beta",
    "adetailer_face": false,
    "adetailer_hand": false,
    "refiner_prompt": "",
    "negative_prompt": "nsfw, naked",
    "adetailer_person": false,
    "guidance_rescale": 1,
    "refiner_strength": 0.8,
    "prepend_preprompt": true,
    "prompt_conjunction": true,
    "adetailer_face_prompt": "",
    "adetailer_hand_prompt": "",
    "adetailer_person_prompt": "",
    "negative_prompt_conjunction": false,
    "adetailer_face_negative_prompt": "",
    "adetailer_hand_negative_prompt": "",
    "adetailer_person_negative_prompt": ""
  },
  "logs": "Balancing prompt's shape\nUsing text to image mode\nUsing seed: 872070416\n  0%|          | 0/30 [00:00<?, ?it/s]\n  7%|▋         | 2/30 [00:00<00:02, 11.35it/s]\n 13%|█▎        | 4/30 [00:00<00:02,  9.37it/s]\n 17%|█▋        | 5/30 [00:00<00:02,  9.04it/s]\n 20%|██        | 6/30 [00:00<00:02,  8.82it/s]\n 23%|██▎       | 7/30 [00:00<00:02,  8.67it/s]\n 27%|██▋       | 8/30 [00:00<00:02,  8.57it/s]\n 30%|███       | 9/30 [00:01<00:02,  8.50it/s]\n 33%|███▎      | 10/30 [00:01<00:02,  8.45it/s]\n 37%|███▋      | 11/30 [00:01<00:02,  8.41it/s]\n 40%|████      | 12/30 [00:01<00:02,  8.38it/s]\n 43%|████▎     | 13/30 [00:01<00:02,  8.37it/s]\n 47%|████▋     | 14/30 [00:01<00:01,  8.36it/s]\n 50%|█████     | 15/30 [00:01<00:01,  8.35it/s]\n 53%|█████▎    | 16/30 [00:01<00:01,  8.35it/s]\n 57%|█████▋    | 17/30 [00:01<00:01,  8.33it/s]\n 60%|██████    | 18/30 [00:02<00:01,  8.32it/s]\n 63%|██████▎   | 19/30 [00:02<00:01,  8.32it/s]\n 67%|██████▋   | 20/30 [00:02<00:01,  8.30it/s]\n 70%|███████   | 21/30 [00:02<00:01,  8.29it/s]\n 73%|███████▎  | 22/30 [00:02<00:00,  8.28it/s]\n 77%|███████▋  | 23/30 [00:02<00:00,  8.28it/s]\n 80%|████████  | 24/30 [00:02<00:00,  8.27it/s]\n 83%|████████▎ | 25/30 [00:02<00:00,  8.25it/s]\n 87%|████████▋ | 26/30 [00:03<00:00,  8.26it/s]\n 90%|█████████ | 27/30 [00:03<00:00,  8.26it/s]\n 93%|█████████▎| 28/30 [00:03<00:00,  8.25it/s]\n 97%|█████████▋| 29/30 [00:03<00:00,  8.24it/s]\n100%|██████████| 30/30 [00:03<00:00,  8.24it/s]\n100%|██████████| 30/30 [00:03<00:00,  8.45it/s]",
  "metrics": {
    "predict_time": 6.209139565,
    "total_time": 6.216409
  },
  "output": [
    "https://replicate.delivery/xezq/iVFQnKpg217hA9xO2Kz0jUi9rXW1ITalN9GEUqE7fZCGgPNKA/0.png"
  ],
  "started_at": "2025-03-20T16:10:45.912269Z",
  "status": "succeeded",
  "urls": {
    "stream": "https://stream.replicate.com/v1/files/bcwr-inosqx67gnkyvlcz3l7qf5dcjhmronkekxepbafrh45ojs2wtwsa",
    "get": "https://api.replicate.com/v1/predictions/fw5b2a2725rm80cnph9b96dp44",
    "cancel": "https://api.replicate.com/v1/predictions/fw5b2a2725rm80cnph9b96dp44/cancel"
  },
  "version": "635cd726c744b1f5cd2e4fc3a801ac1c9943beb6d1ca04fef7300058e78919af"
}

Generated in

6.2 seconds

Tweak it ShareReport View full prediction

This output was created using a different version of the model, aisha-ai-official/anillustrious-v4:635cd726.

Examples

View more examples

Run time and cost

This model costs approximately $0.0048 to run on Replicate, or 208 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 5 seconds.

Readme

This model doesn't have a readme.