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

VeOmni: Scaling Any Modality Model Training with Model-Centric Distributed Recipe Zoo

verl: Volcano Engine Reinforcement Learning for LLMs

Official code for the paper "Compositional Generalization Requires Linear, Orthogonal Representations in Vision Embedding Models"

Official implementation of the Reconstruct Anything Model (RAM)

PiSSA: Principal Singular Values and Singular Vectors Adaptation of Large Language Models(NeurIPS 2024 Spotlight)

Official implementation for the WACV 2026 paper "Deepfake Detection that Generalizes Across Benchmarks".

A comprehensive benchmark of deepfake detection

State-of-the-art 2D and 3D Face Analysis Project

Official implementation of ICML 2025 Oral 🏆 paper "Orthogonal Subspace Decomposition for Generalizable AI-Generated Image Detection".

A Datacenter Scale Distributed Inference Serving Framework

Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.

Achieve state of the art inference performance with modern accelerators on Kubernetes

Towards Pixel-Level VLM Perception via Simple Points Prediction

TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting.

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

Curated collection of AI inference engineering resources — LLM serving, GPU kernels, quantization, distributed inference, and production deployment. Compiled from the AER Labs community.

Model-based design and verification for robotics.

Code at the speed of thought – Zed is a high-performance, multiplayer code editor from the creators of Atom and Tree-sitter.

Pixio: a capable vision encoder dedicated to dense prediction, simply by pixel reconstruction

omo; the best agent harness - previously oh-my-opencode

AirLLM 70B inference with single 4GB GPU

[ICLR 2026] Official implementation for What matters for Representation Alignment: Global Information or Spatial Structure?

The contents of /home/oai/skills in ChatGPT's code interpreter environment

[ICLR 2026] Official PyTorch Implementation of RLP: Reinforcement as a Pretraining Objective

[CVPR2026] Detect Anything via Next Point Prediction

OCR, layout analysis, reading order, table recognition in 90+ languages

This repository is established to store personal notes and annotated papers during daily research.