Research and Production Oriented Speaker Verification, Recognition and Diarization Toolkit

OpenVE-3M: A Large-Scale High-Quality Dataset for Instruction-Guided Video Editing

NextFlow🚀: Unified Sequential Modeling Activates Multimodal Understanding and Generation

HY-World 1.5: A Systematic Framework for Interactive World Modeling with Real-Time Latency and Geometric Consistency

Semantics Lead the Way: Harmonizing Semantic and Texture Modeling with Asynchronous Latent Diffusion

[Preprint 2025] Ditto: Scaling Instruction-Based Video Editing with a High-Quality Synthetic Dataset

Native Multimodal Models are World Learners

Codec for paper: LLaSA: Scaling Train-time and Inference-time Compute for LLaMA-based Speech Synthesis

Official repository for BrickGPT, the first approach for generating physically stable toy brick models from text prompts.

Official PyTorch Implementation of "Diffusion Transformers with Representation Autoencoders"

[NeurIPS'25 Spotlight] Boosting Generative Image Modeling via Joint Image-Feature Synthesis

Image Tokenizer Needs Post-Training

[ICLR 2026] Code for our paper "Next Visual Granularity Generation".

Industry-level video foundation model for unified Text-to-Video (T2V) and Image-to-Video (I2V) generation.

[🚀 ICLR 2026 Oral]NextStep-1: SOTA Autogressive Image Generation with Continuous Tokens. A research project developed by the StepFun’s Multimodal Intelligence team.

Qwen-Image is a powerful image generation foundation model capable of complex text rendering and precise image editing.

Wan: Open and Advanced Large-Scale Video Generative Models

Test-time Scaling for VAR models

Official PyTorch Implementation of "Latent Denoising Makes Good Visual Tokenizers"

Official codebase for "Self Forcing: Bridging Training and Inference in Autoregressive Video Diffusion" (NeurIPS 2025 Spotlight)

Code and dataset link for "DenseWorld-1M: Towards Detailed Dense Grounded Caption in the Real World"

[NeurIPS 2025] Efficient Reasoning Vision Language Models

诺亚盘古大模型研发背后的真正的心酸与黑暗的故事。

[NeurIPS 2025] Geometry Aware Operator Transformer As An Efficient And Accurate Neural Surrogate For PDEs On Arbitrary Domains

Train transformer language models with reinforcement learning.

An official implementation of DanceGRPO: Unleashing GRPO on Visual Generation

🔥 Official impl. of "DetailFlow: 1D Coarse-to-Fine Autoregressive Image Generation via Next-Detail Prediction"

[NeurIPS 2025 D&B] Open-source Multi-agent Poster Generation from Papers