A compact implementation of SGLang, designed to demystify the complexities of modern LLM serving systems.

multilspy is a lsp client library in Python intended to be used to build applications around language servers.

Data transformation framework for AI. Ultra performant, with incremental processing. 🌟 Star if you like it!

Submit stacked diffs to GitHub on the command line

Building the Virtuous Cycle for AI-driven LLM Systems

A powerful coding agent toolkit providing semantic retrieval and editing capabilities (MCP server & other integrations)

Repo2Run is an LLM-based agent that automates environment configuration by generating error-free Dockerfiles for Python repositories.

A Heterogeneous Benchmark for Information Retrieval. Easy to use, evaluate your models across 15+ diverse IR datasets.

RankLLM is a Python toolkit for reproducible information retrieval research using rerankers, with a focus on listwise reranking.

The memory-first coding agent

A next generation Python CMake adaptor and Python API for plugins

Open-source graph database, tuned for dynamic analytics environments. Easy to adopt, scale and own.

Build and publish crates with pyo3, cffi and uniffi bindings as well as rust binaries as python packages

a size profiler for cuda binary

A Lightweight LLM Post-Training Library

A Framework for LLM-based Multi-Agent Reinforced Training and Inference

Multi-Turn RL Training System with AgentTrainer for Language Model Game Reinforcement Learning

A clean, modular SDK for building AI agents with OpenHands V1.

Fast low-bit matmul kernels in Triton

Secure environments for developers and their agents

Command line AI assistant written in Nim

Build, evaluate and train General Multi-Agent Assistance with ease

[NeurIPS '25] GSO: Challenging Software Optimization Tasks for Evaluating SWE-Agents

A high-throughput and memory-efficient inference and serving engine for LLMs

The code repository of the paper: Competition and Attraction Improve Model Fusion

AIRA-dojo: a framework for developing and evaluating AI research agents

Scalable long-context LLM decoding that leverages sparsity—by treating the KV cache as a vector storage system.

Qwen3-Coder is the code version of Qwen3, the large language model series developed by Qwen team, Alibaba Cloud.