Reliable & unreliable messages over UDP. Robust message fragmentation & reassembly. P2P networking / NAT traversal. Encryption.

The agent engineering platform.

The best way to get AI coding agents to solve hard problems in complex codebases.

Open-source, end-to-end platform for evaluating, observing, and improving LLM and AI agent applications. Tracing · Evals · Simulations · Datasets · Gateway · Guardrails. Self-hostable. Apache 2.0.

OpenShell is the safe, private runtime for autonomous AI agents.

The best-benchmarked open-source AI memory system. And it's free.

A fast, helpful, and open-source document parser

⚔ Multiplayer Framework for Node.js

Secure, Fast, and Extensible Sandbox runtime for AI agents.

Python SDK, Proxy Server (AI Gateway) to call 100+ LLM APIs in OpenAI (or native) format, with cost tracking, guardrails, loadbalancing and logging. [Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthr…

The batteries-included agent harness.

The Open Source control plane for self-hosted, BYOC, and on-prem deployments. Everything you need to distribute applications to self-hosted customers out of the box. Supporting Docker Compose, Dock…

AirLLM 70B inference with single 4GB GPU

Multi-Agent Harness for Production AI

All insights extracted from relevant context engineering and memory papers

Privacy architecture for Momo

📑 PageIndex: Document Index for Vectorless, Reasoning-based RAG

👨‍🚀 Turn API specifications into production-ready SDKs, validators, mocks, and more. 20+ plugins. Millions of weekly npm downloads. Used by Vercel, OpenCode, PayPal, AWS, Autodesk, and many more.

Algorithm powering the For You feed on X

rclcpp (ROS Client Library for C++)

Cosmos-Predict2.5, the latest version of the Cosmos World Foundation Models (WFMs) family, specialized for simulating and predicting the future state of the world in the form of video.

[EMNLP'23, ACL'24] To speed up LLMs' inference and enhance LLM's perceive of key information, compress the prompt and KV-Cache, which achieves up to 20x compression with minimal performance loss.

Fastest enterprise AI gateway (50x faster than LiteLLM) with adaptive load balancer, cluster mode, guardrails, 1000+ models support & <100 µs overhead at 5k RPS.

Specification and documentation for the Universal Commerce Protocol (UCP)

Memory layer for AI Agents. Replace complex RAG pipelines with a serverless, single-file memory layer. Give your agents instant retrieval and long-term memory.

Build, run and scale AI agents like API and microservices - observable,auditable and identity-aware from day one.

Continued development of pufferdrive

Puffing up reinforcement learning

[MLsys2026]: RAG on Everything with LEANN. Enjoy 97% storage savings while running a fast, accurate, and 100% private RAG application on your personal device.