Explore the directory

A Rust-built control plane and data plane for managing MCP services, letting a resource-constrained host attach only the data plane and call already-shared MCP services with no local maintenance.

• MCP servers

A Java 8+ agentic development toolkit that unifies access to more than a dozen large-language-model platforms behind one API, with built-in tool calling, RAG, MCP client and server support, and a long-running agent harness.

• Agent frameworks

Ivy is a C# framework for building full-stack, reactive web applications entirely in C# with React-style components and hooks, without a separate frontend or hand-written HTML, CSS or JavaScript.

• Web frameworks

A NeurIPS 2026 paper and reference implementation showing that standard vision-language models can be trained to do object-level 3D understanding, depth estimation, pixel correspondence and camera pose estimation without special architectures or losses.

• Models
• Papers

The companion repository for a four-week course on deploying generative and agentic AI at scale, organized week by week with guides and space for students to share their own deployed projects.

• Tutorials

A CLI coding agent for long-horizon tasks that isolates execution noise from the model's context, aimed especially at running well with local models in a small context window.

• Coding agents

A community-built, vLLM-alike inference engine written entirely in C++, with continuous batching, paged key-value cache and RadixAttention, aiming for the same throughput as vLLM with a much smaller install.

• Inference

A local-first, evidence-first AI coding terminal that constrains a language model to read real code, verify its own changes, and disclose uncertainty instead of asserting that work is complete without proof.

• Coding agents

A production-oriented Next.js starter with authentication, a database, AI SDK integration and shadcn/ui components, plus written agent instructions and skills so a coding agent can plan, implement, review and verify a new application.

• Agent skills
• Web frameworks

APM Studio is a local GUI for the Agent Package Manager ecosystem that imports agents, skills, prompts, hooks and MCP configurations from GitHub and syncs selected packages out to Codex, Claude, Cursor, Gemini and OpenCode.

• Agent skills

A Python package that automatically compresses large language models, detecting available GPU memory, choosing a bit-width per layer, quantizing with error correction, and evaluating the result in one command.

• Models

An AI coding workspace built around a durable, crash-safe agent core, recording every prompt, tool call and event before it runs so a session can resume exactly where it stopped after a crash or restart.

• Coding agents

A design skill for coding agents that gives them design judgment for UI, presentations and graphics, so generated interfaces look considered rather than like generic AI-produced layouts.

• Agent skills
• Design tools

coding-plan-assistant is a browser extension and macOS menu-bar app that tracks AI subscription plans and token usage across several Chinese AI platforms and surfaces deals, reviews and AI news.

• Monitoring
• Desktop utilities

vLLM Router is a high-performance request router for large-scale vLLM deployments, providing cache-aware and other load-balancing algorithms plus prefill and decode disaggregation and Kubernetes-native service discovery.

• Inference
• Networking

A Rust framework for building composable AI agent runtimes that combine models, tools, memory and other agents into a single, type-safe, asynchronous runtime.

• Agent frameworks

The TPU inference backend for vLLM, unifying JAX and PyTorch model execution under a single lowering path so PyTorch models can run performantly on TPU hardware without extra code changes.

• Inference

A searchable, filterable gallery of well-designed Figure 1 and teaser images from top machine-learning conference papers between 2023 and 2026, meant as visual reference for designing a paper's own opening figure.

• Design tools
• Papers

A terminal orchestrator that supervises headless Claude Code agents through a gated plan-to-merge pipeline, with human review at each design phase and guardrails the agents cannot bypass during the build.

• Coding agents

A curated, evolving guide to building production-ready AI agents and multi-agent systems, covering foundations, frameworks, system design patterns, real-world use cases and interview preparation.

• Tutorials
• Curated lists

An offensive security framework for agentic infrastructure that captures credentials and configuration from a compromised host, fingerprints reachable AI services such as MCP servers and model gateways, and builds an attack graph.

• Security

A local LLM inference server for Apple Silicon and NVIDIA GPUs that exposes an OpenAI-compatible chat endpoint, with custom Metal and CUDA kernels tuned per model family for exact, faster decoding.

• Inference

A runtime that checks an AI agent's tool calls against deterministic rules before they run, in a fraction of a millisecond per check and without calling any model.

• AI evals
• Security

A reinforcement-learning framework that trains language-model agents inside a central-agent multi-agent harness without flattening execution into a single token sequence, routing gradients to action and argument tokens separately.

• Agent frameworks
• Models