HY-World 2.0: A Multi-Modal World Model for Reconstructing, Generating, and Simulating 3D Worlds

VideoFlexTok: Flexible-Length Coarse-to-Fine Video Tokenization

Tempo: Small Vision-Language Models are Smart Compressors for Long Video Understanding

Code for the Boxer research paper

"Single-image Layer Decomposition for Anime Characters" (SIGGRAPH 2026, Conditionally Accepted)

High-Quality Voice Cloning TTS for 600+ Languages

Penguin-VL: Exploring the Efficiency Limits of VLM with LLM-based Vision Encoders [Technical Report]

Sharp Monocular View Synthesis in Less Than a Second

CVPR 2025🎉 Offical repository for "TAG-MoE: Task-Aware Gating for Unified Generative Mixture-of-Experts"

MegaFlow: Zero-Shot Large Displacement Optical Flow

PixelSmile: Fine-grained facial expression editing with continuous control, reduced semantic entanglement, and strong identity preservation.

AutoGaze automatically removes redundant patches in a video, reducing #tokens in ViT/MLLM by 4x-100x.

Mount Hugging Face Buckets and repos as local filesystems. No download, no copy, no waiting.

[ICLR 2026] RF-DETR is a real-time object detection and segmentation model architecture developed by Roboflow, SOTA on COCO, designed for fine-tuning.

Qianfan-VL: Domain-Enhanced Universal Vision-Language Models

PyTorch code and models for VJEPA2 self-supervised learning from video.

Any2Full: Prompting Depth Anything for Depth Completion in One Stage

DVD: Deterministic Video Depth Estimation with Generative Priors

Helios: Real Real-Time Long Video Generation Model

CVPR 2026 - Generative Quanta Image Reconstruction

[CVPR 2026🔥] 🧑‍🎨 OmniLottie, an open-sourced multi-modal instructed vector animation generator that produces Lottie JSONs.

Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflo…

Real-time text-to-speech with Qwen3-TTS

[CVPR 2026] Official code and models for Video Encoder-only Mask Transformer (VidEoMT).

Give your agents the power of the Hugging Face ecosystem

Official implementation of "Rolling Sink: Bridging Limited-Horizon Training and Open-Ended Testing in Autoregressive Video Diffusion"

A real-time and multilingual speech translation model

OpenResearcher: A Fully Open Pipeline for Long-Horizon Deep Research Trajectory Synthesis

Official inference code for SoulX-Singer: Towards High-Quality Zero-Shot Singing Voice Synthesis