[ICLR 2026] PixNerd: Pixel Neural Field Diffusion

Official implementation of AsymFlow, pi-Flow, GMFlow

Flow Map OPD for AnyStep Video Diffusion

Awesome Visual Tokenizers/Autoencoders

[ICLR 2026] PixNerd: Pixel Neural Field Diffusion

[CVPR 2025 Highlight] Towards Enhanced Image Inpainting: Mitigating Unwanted Object Insertion and Preserving Color Consistency

Matrix imputation leveraging external covariance structure information

Official implementation of BLIP3o-Series

the official repo for "D-AR: Diffusion via Autoregressive Models"

[CVPR 2026] DDT: Decoupled Diffusion Transformer

Code for: "Long-Context Autoregressive Video Modeling with Next-Frame Prediction"

Helpful tools and examples for working with flex-attention

[CVPR 2025] Open-source, End-to-end, Vision-Language-Action model for GUI Agent & Computer Use.

[NeurIPS 2024] Exploring DCN-like Architectures for Fast Image Generation with Arbitrary Resolution

Code for [CVPR 2025] ROICtrl: Boosting Instance Control for Visual Generation

FQGAN: Factorized Visual Tokenization and Generation

[ICLR & NeurIPS 2025] Repository for Show-o series, One Single Transformer to Unify Multimodal Understanding and Generation.

[NeurlPS 2024] One Token to Seg Them All: Language Instructed Reasoning Segmentation in Videos

[ICLR 2025] Binary Spherical Quantization + [CVPR 2026] Leech Spherical Quantization

Reference implementation for DPO (Direct Preference Optimization)

VideoLLM-online: Online Video Large Language Model for Streaming Video (CVPR 2024)

[ECCV 2024] DragAnything: Motion Control for Anything using Entity Representation

[CVPR 2024] Panda-70M: Captioning 70M Videos with Multiple Cross-Modality Teachers

Strong and Open Vision Language Assistant for Mobile Devices

A family of lightweight multimodal models.

TensorDict is a pytorch dedicated tensor container.

HOSNeRF: Dynamic Human-Object-Scene Neural Radiance Fields from a Single Video

MLX: An array framework for Apple silicon