- Suzhou. China
- https://ustack.io
Highlights
Starred repositories
Any model. Any hardware. Zero compromise. Built with @ziglang / @openxla / MLIR / @bazelbuild
agent-sandbox enables easy management of isolated, stateful, singleton workloads, ideal for use cases like AI agent runtimes and reinforcement learning (RL).
Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities
Low-level unprivileged sandboxing tool used by Flatpak and similar projects
slime is an LLM training framework for RL Scaling.
AI agents running research on single-GPU nanochat training automatically
A Python library for extracting structured information from unstructured text using LLMs with precise source grounding and interactive visualization.
Use this skill to enable Claude Code to communicate directly with your Google NotebookLM notebooks. Query your uploaded documents and get source-grounded, citation-backed answers from Gemini. Featu…
TurboDiffusion: 100–200× Acceleration for Video Diffusion Models
[ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-to-end metrics across language, image, and video models.
Own your AI. The open-source agent harness for macOS. Run local or cloud models with persistent memory, skills, tools, and autonomous execution. Built in Swift.
Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4.1, MiniMax-H3, Gemma 4, FLUX and more.
My learning notes for ML SYS.
🚀 EvoAgentX: Building a Self-Evolving Ecosystem of AI Agents
verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework
Python SDK for AI agent monitoring, LLM cost tracking, benchmarking, and more. Integrates with most LLMs and agent frameworks including CrewAI, Agno, OpenAI Agents SDK, Langchain, Autogen, AG2, and…
LMCache: Supercharge Your LLM with the Fastest KV Cache Layer
🤯 LobeHub - Super-Agents on Your Team. Add super-agents to the chats your team already uses.They take on work by themselves and get better every week.
SGLang is a high-performance serving framework for large language models and multimodal models.
微舆:人人可用的多Agent舆情分析助手,打破信息茧房,还原舆情原貌,预测未来走向,辅助决策!从0实现,不依赖任何框架。
real time face swap and one-click video deepfake with only a single image
Wan: Open and Advanced Large-Scale Video Generative Models
Virtualized Elastic KV Cache for Dynamic GPU Sharing and Beyond
The Future of Data Engineering — A CLI SQL client for the modern data stack, enabling AI-native context engineering for data.
The open source AI engineering platform for agents, LLMs, and ML models. MLflow enables teams of all sizes to debug, evaluate, monitor, and optimize production-quality AI applications while control…
The world's fastest open query engine for sub-second analytics both on and off the data lakehouse. With the flexibility to support nearly any scenario, StarRocks provides best-in-class performance …