ComfyUI-Tagger
ComfyUI node integration for the Camie tagger v2 model from Hugging Face for image tagging. (Description by CC)
Nodes (4)
The 70k-tag anime tagger that was trained on one RTX 3060
The one that tells you which franchise the image is from
Set math for taggers, no model download required
The aspect-ratio-aware ViT-H tagger nobody's heard of yet
PixAI Tagger
<img width="1223" height="592" alt="image" src="https://github.com/user-attachments/assets/94f67a9b-dea4-4663-9140-dc0856740903" />Official source: https://huggingface.co/pixai-labs/pixai-tagger-v0.9
Download three files (any path supported):
model_v0.9.pth
char_ip_map.json
tags_v0.9_13k.json
Many thanks to PixAI Labs
Camie Tagger
<img width="1004" height="836" alt="image" src="https://github.com/user-attachments/assets/44f32879-240b-4152-a25e-99eff3055630" />Official source: https://huggingface.co/Camais03/camie-tagger-v2
Download two files (any path supported):
camie-tagger-v2.onnx
camie-tagger-v2-metadata.json
Many thanks to Camais03
Taggerine Tagger
<img width="1075" height="808" alt="image" src="https://github.com/user-attachments/assets/e4d5b445-1349-4afc-a25d-d69f679d72ea" />Official source: https://huggingface.co/lodestones/taggerine
Download two files (any path supported):
tagger_proto.safetensors
tagger_vocab_with_categories_and_alias_updated.json
Many thanks to lodestones
Tag Combiner
You can pass 2-4 tag results to this node and combine them using subtraction or merging of unique tags.
<img width="1322" height="764" alt="image" src="https://github.com/user-attachments/assets/ae6e406e-8b51-4ed0-b4a0-49fc50bb867e" />