autoresearch‑claude‑code is a Python framework that runs budget‑constrained auto‑research loops with Claude Code or Codex agents. You define a goal, a deterministic scorer, and files the agent may edit; the system creates a Git branch, measures each change, and automatically commits only statistically‑significant improvements while respecting run‑time and turn limits.

A tool for automating the and evaluation of multiple LLMs, prompts, and inference parameters via Ollama to find the optimal configuration for a specific task.

DR Tulu is an open‑source 8 B‑parameter model and toolkit for building research‑oriented agents that can browse, retrieve scholarly information, and generate long‑form answers. The repo provides an agent library with async tool‑calling, RL training (GRPO + evolving rubrics), SFT pipelines, evaluation scripts for multiple research benchmarks, and a CLI demo that runs on 1‑2 GPUs.

An open-source Gymnasium environment for simulating building energy coordination and demand response using multi-agent reinforcement learning.

A unified framework for Scientific Machine Learning (SciML) providing implementations of PINNs, DeepONets, and KANs for studying complex physical systems.

A collection of Python scripts to test OpenAI-compatible model endpoints to verify if the model being served matches its label.

A high-performance inference engine that uses SVDQuant to enable 4-bit quantization for diffusion models, significantly reducing VRAM usage and increasing inference speed.

A tuned OpenCL BLAS library that provides high-performance linear algebra routines for a wide variety of GPUs and accelerators.

A multi-tenant Postgres platform that decouples storage from compute to help B2B SaaS applications scale securely and efficiently with integrated vector database support.

A reconciliation-based coding agent that prevents context rotting by using fresh contexts and file-system memory to iteratively align a codebase with a specified goal.

An AI-native software testing platform that uses a single YAML-based declarative language to automate API, Web, mobile, and LLM tests without requiring traditional code scripts.

A web application for browsing, searching, and exporting chat histories and Composer logs from the Cursor AI editor.

A tool for visualizing and exploring unstructured text and image datasets by embedding them into latent spaces, projecting them to 2D, and using LLMs to label clusters.

A backend for ONNX that converts ONNX models into TensorFlow models, enabling them to be run using the TensorFlow execution engine.

A cost-efficient routing system that replaces expensive LLM classification calls with lightweight traditional ML surrogates for easy inputs, maintaining teacher-level parity.

A unified runtime for LLM agents to execute structured tool-call chains, featuring deep integration with the Solana blockchain and support for the Model Context Protocol (MCP).

A unified and lightweight framework for the curation, packaging, loading, and evaluation of massive AI datasets, supporting multiple formats like LMDB and Pickle.

A flexible grid-based environment for Partially-Observable Multi-Agent Pathfinding (PO-MAPF), designed for testing decentralized navigation and collision avoidance in multi-agent systems.

A provider-agnostic PHP SDK that provides a uniform API for interacting with various generative AI models for text, image, and embedding generation.

An unofficial Dart/Flutter SDK providing typed clients for the OpenAI API and compatible providers, supporting all major platforms including web.

A host-agnostic safety middleware that protects tool-using AI agents from executing risky commands and leaking sensitive data.

A GenAI-powered, codeless test automation platform that enables QA teams to generate and maintain tests for web, mobile, and desktop apps without writing code.

A highly customizable terminal status line for Claude Code that displays real-time metrics, cost tracking, and environment status across 1-9 configurable lines.

A comprehensive SaaS blueprint for B2B/B2C products featuring a .NET 10 and React 19 stack, integrated Stripe billing, and an AI-native multi-agent development workflow.

A lightweight CPU and GPU library for computing nonuniform Fast Fourier Transforms (NUFFT) in one to three dimensions to a specified precision.

LAVA is a tool for automatically injecting synthetic but realistic vulnerabilities into software to create ground truth corpora for evaluating bug-finding tools.

A cross-platform Directed Acyclic Graph (DAG) framework in C++ and Python that allows developers to build complex task pipelines with automatic dependency management and parallel execution.

A Python package for the exploratory analysis of Bayesian models, providing tools for posterior analysis, model checking, and diagnostics.

Neanderthal is a Clojure library for high-performance matrix and linear algebra computations utilizing optimized native BLAS and LAPACK routines on CPU and GPU.

An R package for assessing, comparing, and testing the quality of statistical regression models through fit indices and diagnostic checks.

An interactive, high-performance data table for Jupyter notebooks that streamlines exploratory data analysis with features like autocleaning, histograms, and lazy loading.

A bio-inspired computational framework using trinary logic and continuous-time simulation to enable real-time learning and structural plasticity.

A modern git interface for macOS and Linux designed to help developers oversee and manage the high volume of code changes generated by AI coding agents.

An open-source static code security analysis system that uses AST graphs and taint analysis to detect vulnerabilities across 14 different programming languages.

A Python toolbox for visualizing high-dimensional data through an integrated pipeline of dimensionality reduction, clustering, and advanced 2D/3D plotting.

A Node.js client library for integrating IBM Watson AI services, providing simplified authentication and API access for tools like Watson Assistant and Discovery.

A high-level Bayesian model-building interface for Python that simplifies fitting mixed-effects and linear models using PyMC.

A Julia library for computing derivatives, gradients, Jacobians, and Hessians using forward-mode automatic differentiation.

A header-only C++ library providing thin, modern wrappers for NVIDIA's CUDA Runtime, Driver, NVRTC, and NVTX APIs to simplify GPU programming and resource management.

A multidimensional array library for Kotlin that provides type-safe N-dimensional arrays, linear algebra, and statistics across multiple platforms.

A cross-platform log viewer that parses and analyzes logs from diverse sources and integrates with AI assistants via the Model Context Protocol (MCP).

A lightweight Vue 3 prototype template for building AI chat interfaces, featuring multi-model API integration and streaming rich-text rendering.

A high-performance Python/C++ library for the fast simulation of large, noisy, or parametric quantum circuits, supporting both CPU multi-threading and GPU acceleration.

A framework for building and evaluating information retrieval pipelines, supporting sparse and dense retrieval, neural reranking, and RAG use-cases.

A header-only C++ template library providing custom arithmetic plug-in types to optimize AI and HPC algorithms through mixed-precision and alternative number formats.

A Julia package that enables GPU programming on macOS M-series chips using the Metal framework, providing both high-level array abstractions and low-level API access.

A fast, parallelized sparse matrix multiplication library that integrates top-n result selection to reduce memory usage and improve performance when matching large feature vectors.

An Ansible-based framework for automating the setup and benchmarking of distributed AI clusters using tools like llama.cpp and distributed-llama.