A Korean sentiment classification model based on KoELECTRA fine - tuning, capable of identifying six emotions: anger, happiness, anxiety, embarrassment, sadness, and heartache.

Natural Language ProcessingTransformersKorean

HeAR is a health acoustic foundation model focused on representation learning of non-semantic respiratory sounds such as coughs and breathing patterns, applicable for health monitoring and disease screening.

Audio ProcessingTransformersEnglish

A LoRA trained based on the Wan2.1 14B I2V 480p model, specifically designed for generating videos with characters laughing heartily.

MultimodalDiffusersEnglish

LoRA trained based on Wan2.1 14B I2V 480p model, specifically for generating videos of people laughing heartily

MultimodalDiffusersEnglish

EMOVA is an end-to-end omnipotent modality large language model that supports vision, hearing, and speech functions, enabling multimodal understanding and generation without relying on external models.

MultimodalTransformersMultiple Languages

An audio classification model fine-tuned based on MIT/ast-finetuned-audioset-10-10-0.4593, specializing in heart sound classification tasks, achieving 96.95% accuracy on the validation set.

Audio ProcessingTransformers

This is a text-to-image generation model based on LoRA technology, specifically designed to generate realistic and high-quality photos of Amber Heard. The model is based on the FLUX.1-dev architecture and activates the image generation function through the specific trigger word 'ohwx'.

Computer VisionDiffusers

This model performs simultaneous segmentation and classification tasks on chest X-rays, including lung/heart segmentation, position recognition, and age/gender prediction.

Computer VisionTransformers

A Hearthstone card generation model fine-tuned based on Stable-Diffusion v2.1, capable of generating Hearthstone-style card images from text prompts.

MultimodalDiffusersEnglish