TurboDiffusion: 100–200× Acceleration for Video Diffusion Models

FloodDiffusion: Tailored Diffusion Forcing for Streaming Motion Generation

Trim videos in the terminal

The official implementation of The paper "Are We Ready for RL in Text-to-3D Generation? A Progressive Investigation"

V-RGBX: Video Editing with Accurate Controls over Intrinsic Properties

PersonaLive! : Expressive Portrait Image Animation for Live Streaming

`scm-record` is a UI component to interactively select changes to include in a commit. It's meant to be embedded in source control tooling.

A real-time streaming conversational video system that transforms text interactions into continuous, high-fidelity video responses using autoregressive diffusion.

Implementation of "Live Avatar: Streaming Real-time Audio-Driven Avatar Generation with Infinite Length"

[NeurIPS 2025] Wan-Move: Motion-controllable Video Generation via Latent Trajectory Guidance

ConceptAttention: A method for interpreting multi-modal diffusion transformers.

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

A fast partial replacement for the codemod tool. Assists with large-scale codebase refactors via regex-based find and replace with human oversight and occasional intervention.

Low-level packet manipulation library for Nim that lets you build and parse Ethernet, IPv4, TCP, UDP, and ICMP as structured data instead of raw bytes.

🚀2.3x faster than MinIO for 4KB object payloads. RustFS is an open-source, S3-compatible high-performance object storage system supporting migration and coexistence with other S3-compatible platfor…

⚡Batch Face Processing for Fast Modern Research, including face detection, face alignment, face reconstruction, head pose estimation, face parsing

Reward Forcing: Efficient Streaming Video Generation with Rewarded Distribution Matching Distillation

ViSAudio: End-to-End Video-Driven Binaural Spatial Audio Generation

Neo is a new package manager for Nim, aiming to be fast and maintainable

[WACV 2026] LASER: Lip Landmark Assisted Speaker Detection for Robustness official implemntation

Optimized Whisper models for streaming and on-device use

SteadyDancer: Harmonized and Coherent Human Image Animation with First-Frame Preservation

The next-generation file converter. Open source, fully local* and free forever.

s&box is a modern game engine, built on Valve's Source 2 and the latest .NET technology, it provides a modern intuitive editor for creating games

Official inference repo for FLUX.2 models

[AAAI 2026] Playmate2: Training-Free Multi-Character Audio-Driven Animation via Diffusion Transformer with Reward Feedback

Swift Package to implement a transformers-like API in Swift