Models

This is an ONNX-exported Whisper model fine-tuned for language identification from audio inputs.

This is an image classification model for tagging real-world photos with ratings, characters, and general tags.

This is an image classification model specifically designed for tagging anime-style content with ratings, characters, and general tags.

This is an ONNX-optimized Whisper model fine-tuned for identifying 45 different languages from audio input.

This is an ONNX-optimized Whisper model fine-tuned for identifying 45 different languages from audio input.

This is an ONNX-optimized image classification model for multilabel tagging of anime-style artwork with 13,461 tags across general and character categories.

This repository provides specialized YOLOv8-based object detection models for accurately detecting anime character faces in images.

A collection of deep learning models for classifying anime images into categories including 3D renders, anime screenshots, manga comics, illustrations, and non-painting content.

This repository provides optimized ONNX models for anime image classification and tagging with enhanced embedding capabilities.

This repository provides YOLO-based object detection models specifically trained to identify key elements in manga images including bodies, faces, frames, and text.

This repository provides ONNX-format PaddleOCR models for multilingual optical character recognition with text detection and recognition capabilities.

CCIP is a contrastive learning model that calculates visual similarity between anime characters in single-character images for zero-shot character recognition and clustering.

This repository provides YOLOv8-based object detection models specifically trained to identify censored anatomical regions in anime images.

This repository provides deep learning models for classifying anime images into seven aesthetic quality categories.

This repository contains optimized YOLO face detection models for real-time object detection applications.

A collection of specialized computer vision models for anime image processing including classification, detection, tagging, and enhancement tasks.

This repository provides specialized YOLOv8 object detection models for detecting anime-style human characters in images.

This repository contains optimized DBNet and DBNet++ models for text detection in images, supporting multiple backbone architectures and trained on standard text detection benchmarks.

This repository contains experimental SigLIP models for zero-shot image classification and image-text similarity tasks, specifically optimized for anime artwork analysis.

This repository provides ONNX models for face detection, recognition, and analysis with high accuracy across multiple benchmark datasets.

This repository contains ONNX-optimized models for zero-shot character similarity classification and image feature extraction.

This repository provides multiple deep learning models for classifying images as either anime-style artwork or real photographs.

This repository provides specialized YOLO-based object detection models for accurately detecting anime character heads in images.

This repository provides ONNX-optimized deep learning models for automated tagging and classification of anime-style images across thousands of Danbooru-style attributes.

This repository provides YOLO12 object detection models specifically trained for detecting text blocks in anime scenes.

This repository provides deep learning models for image restoration tasks including noise reduction, deblurring, and general image enhancement.

This repository provides deep learning models for classifying anime art styles by decade from the 1970s to 2020s.

This repository contains ONNX-format super-resolution models for upscaling anime-style images with 2x and 4x scaling capabilities.

This repository contains image classification models for predicting anime content ratings (safe, r15, r18) using various neural network architectures.

This repository provides specialized object detection models for detecting hands in anime-style images.

A collection of YOLOv8-based object detection models specialized for detecting eyes in anime-style images.

This repository contains YOLOv8 object detection models specialized for detecting anatomical features and artistic elements in anime-style artwork.

This repository provides ONNX-format DINOv2 vision transformer models for efficient image feature extraction across multiple model sizes.

This repository provides optimized ONNX models for efficient nudity detection and NSFW content recognition in images using YOLOv8-based object detection.

This repository provides specialized object detection models for identifying anime half-body characters in images.

This repository contains image classification models for automatically detecting anime character hair length from images.

This repository contains image classification models that categorize anime artwork into polished, rough, or monochrome completeness levels.

This model repository provides classification models for detecting corrupted anime-style stable diffusion generated images.

This model denormalizes WD14 tagger embeddings to restore accurate image tag predictions while preserving cosine similarity properties.

This repository provides pre-built FAISS indices for efficient similarity search and retrieval of anime-style images across multiple popular anime image databases.

A comprehensive collection of pretrained YOLO object detection models for identifying 80 different object categories in images.

This repository provides ONNX-exported CLIP models for efficient zero-shot image classification and cross-modal retrieval tasks.

This repository provides ONNX-exported SigLIP models for efficient zero-shot image classification tasks.

A specialized YOLOv8 model optimized for efficient person detection in images.

This model repository provides multiple image classification models for detecting AI-generated anime artwork versus human-drawn illustrations.

This repository contains multiple image classification models specifically trained to categorize anime character portraits into person, halfbody, and head classes.

This repository contains image classification models for detecting anime character skin tones across four categories: dark, light, pale, and colored.

This model repository provides specialized image classification models for automatically identifying hair colors of anime characters.

This repository provides multiple deep learning models for classifying anime images into content rating categories including general, sensitive, questionable, and explicit.

This repository contains image classification models for distinguishing between furry and non-furry anime-style artwork.

This repository provides specialized image classification models for automated tagging of idol images with content ratings and character attributes.

This repository provides YOLO-based face detection models that work on both anime and real photographs with high accuracy.

This repository provides YOLO-based object detection models for accurate head detection in both anime and real photographic images.

A collection of YOLO-based object detection models for detecting persons in both anime and real photographic images.

This model provides automated tagging and classification of anime-style images using deep learning.

This repository provides multiple image classification models for detecting monochrome anime artwork with high accuracy.

This model performs image classification to identify horn types in anime character designs.

This repository contains image classification models for detecting anime character age categories to assist with content moderation.

This repository contains computer vision models for classifying anime character eye colors from images.

This repository provides image classification models specifically trained to identify character ear types in anime artwork.

This repository provides image classification models for categorizing anime character images into four types: vision, imagery, halfbody, and face.

This is an ONNX-format language detection model that can identify 95 different languages from audio input.

This repository contains ONNX-exported computer vision models from the TIMM library for image classification tasks.

This repository contains experimental MLP models for classifying images as monochrome or normal using DeepGHS's tagger attachment framework.

This is a lightweight image classification model for predicting content ratings (r15, r18, safe) on Sankaku Complex images.

This repository provides ONNX-exported versions of the Camie Tagger model for comprehensive image tagging and classification with 70,527 distinct tags.