Structured Defect Grounding
Ground localized defects in AI-generated images
None defined yet.
Ground localized defects in AI-generated images
Real-time one-step generative video restoration
Speech-to-text transcription with MOSS-Transcribe-preview-2B
Phone-use GUI agent that predicts actions from screenshots
Enhance text-to-image with the Krea2 Enhancer LoRA on Turbo
Krea-2-Turbo with IdeoKrea style LoRA demo
Multilingual speech recognition in 30+ languages
Video understanding with InternVideo3 MΒ²LA multimodal LLM
Unified multimodal video generation and editing (5B)
Zero-shot multilingual NER with GLiNER on LFM2.5-350M
Semantic-First Diffusion text-to-image in 4 steps
PP-OCRv6 medium recognition β text recognition from images
Detect text regions in images with PP-OCRv6 medium
3D-aware object insertion with the DIRECT model (ICML 2026)
Multimodal bio foundation model for molecules & proteins
ASR for 52 languages and dialects
Infinite video generation with learnable evolving memory
Add Foley sound effects to silent videos with LTX-2.3
Slack interactivity endpoint for the HuggingDemos queue
Dialectical reasoning for ICU mortality risk prediction
Action-conditioned video world model demo
Tiny any-to-any multimodal GPT (text + image) prototype demo
Distractor-free 2D enhancer for radiance field renderings
Vision-driven VLM for image and video understanding