Efficient and general syntactical decoding for Large Language Models

Agentless🐱: an agentless approach to automatically solve software development problems

We Have No Moat, And Neither Does Devin

A programming framework for agentic AI. Discord: https://aka.ms/autogen-dc. Roadmap: https://aka.ms/autogen-roadmap

A code-first agent framework for seamlessly planning and executing data analytics tasks.

🐶 Kubernetes CLI To Manage Your Clusters In Style!

Symbolic stack CFG generator for EVM

CodeQL: the libraries and queries that power security researchers around the world, as well as code scanning in GitHub Advanced Security

LDB: A Large Language Model Debugger via Verifying Runtime Execution Step by Step

CodiumAI Cover-Agent: An AI-Powered Tool for Automated Test Generation and Code Coverage Enhancement! 💻🤖🧪🐞

DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model

🙃 A delightful community-driven (with 2,300+ contributors) framework for managing your zsh configuration. Includes 300+ optional plugins (rails, git, macOS, hub, docker, homebrew, node, php, python…

Test Server for R2E

R2E Project Website

A minimal GPU design in Verilog to learn how GPUs work from the ground up

Typed argument parser for Python

Pycopy - a minimalist and memory-efficient Python dialect. Good for desktop, cloud, constrained systems, microcontrollers, and just everything.

Dead simple pool implementation for rust with async-await

paste image from clipboard to markdown/asciidoc directly!

Datalog with support for SMT queries and first-order functional programming

A modern Python package and dependency manager supporting the latest PEP standards

SWE-agent takes a GitHub issue and tries to automatically fix it, using GPT-4, or your LM of choice. It solves 12.47% of bugs in the SWE-bench evaluation set and takes just 1 minute to run.

llmstep: [L]LM proofstep suggestions in Lean 4.

Retrieval-Augmented Theorem Provers for Lean

A project structure aware autonomous software engineer aiming for autonomous program improvement. Resolved 30.67% tasks (pass@1) in SWE-bench lite with each task costs less than $0.7.

Ensures that all commits are following the conventional-commits standard.