W

WD EVA02-Large Tagger v3

Tagger

No ratings yet(0)
Sign in to vote

WD EVA02-Large Tagger v3 is an image generation model (tagger) on Arli AI. It supports image-to-image, inpainting, ControlNet, IP-Adapter. Generate images through the SDNext-compatible Arli AI image API with unlimited requests and zero-log privacy.

View on Hugging FaceBack to Image Models

Hourly Usage

Model Information

Recommended Prompt

N/A

Model Type

tagger

Model Class

Tagger

Recommended Sampler

Euler a

Recommended Steps

20

Recommended CFG

6

CLIP Skip

1

Requests (24h)

1

Requests (7d)

90

Capabilities

Image to image, Inpainting, ControlNet, IP-Adapter

Extra Info

N/A

Added At

8/4/2026

Recommended Negative Prompt

lowres, bad anatomy, bad hands, text, error, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry

Use WD EVA02-Large Tagger v3 with the Arli AI API

Generate images with WD EVA02-Large Tagger v3 through the OpenAI-style image endpoint. Send sd_model_checkpoint as the model name and the API renders the result without logging your prompts.

cURL

curl -X POST https://api.arliai.com/v1/txt2img \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $ARLIAI_API_KEY" \
  -d '{
    "sd_model_checkpoint": "WD EVA02-Large Tagger v3",
    "prompt": "a red fox in fresh snow, cinematic light",
    "negative_prompt": "lowres, bad anatomy, watermark",
    "steps": 20,
    "sampler_name": "Euler a",
    "cfg_scale": 6,
    "width": 1024,
    "height": 1024
  }'

Python

import requests

response = requests.post(
    "https://api.arliai.com/v1/txt2img",
    headers={"Authorization": "Bearer YOUR_ARLIAI_API_KEY"},
    json={
        "sd_model_checkpoint": "WD EVA02-Large Tagger v3",
        "prompt": "a red fox in fresh snow, cinematic light",
        "steps": 20,
        "sampler_name": "Euler a",
        "cfg_scale": 6,
        "width": 1024,
        "height": 1024,
    },
)
print(response.json())

See the image generation docs for every parameter, or read the quick start guide to make your first request.

license: apache-2.0 library_name: timm

WD EVA02-Large Tagger v3

Supports ratings, characters and general tags.

Trained using https://github.com/SmilingWolf/JAX-CV.
TPUs used for training kindly provided by the TRC program.

Dataset

Last image id: 7220105
Trained on Danbooru images with IDs modulo 0000-0899.
Validated on images with IDs modulo 0950-0999.
Images with less than 10 general tags were filtered out.
Tags with less than 600 images were filtered out.

Validation results

v1.0: P=R: threshold = 0.5296, F1 = 0.4772

What's new

Model v1.0/Dataset v3:
More training images, more and up-to-date tags (up to 2024-02-28).
Now timm compatible! Load it up and give it a spin using the canonical one-liner!
ONNX model is compatible with code developed for the v2 series of models.
The batch dimension of the ONNX model is not fixed to 1 anymore. Now you can go crazy with batch inference.
Switched to Macro-F1 to measure model performance since it gives me a better gauge of overall training progress.

Runtime deps

ONNX model requires onnxruntime >= 1.17.0

Inference code examples

For timm: https://github.com/neggles/wdv3-timm
For ONNX: https://huggingface.co/spaces/SmilingWolf/wd-tagger
For JAX: https://github.com/SmilingWolf/wdv3-jax

Final words

Subject to change and updates. Downstream users are encouraged to use tagged releases rather than relying on the head of the repo.