llm
AI for Large Codebases
The first AI-native code platform that gives large projects compiler-grade context, so LLMs stay useful even when your repo spans millions of lines.
Brokk (the Norse god of the forge) is the first code assistant that understands code semantically, not just as chunks of text. Brokk is designed to allow LLMs to work effectively on large codebases that cannot be jammed entirely into working context.
A Flexible Framework for Experiencing Cutting-edge LLM Inference Optimizations.
KTransformers, pronounced as Quick Transformers, is designed to enhance your 🤗 Transformers experience with advanced kernel optimizations and placement/parallelism strategies.
KTransformers is a flexible, Python-centric framework designed with extensibility at its core. By implementing and injecting an optimized module with a single line of code, users gain access to a Transformers-compatible interface, RESTful APIs compliant with OpenAI and Ollama, and even a simplified ChatGPT-like web UI.
Semantic unit testing is a testing approach that evaluates whether a function's implementation aligns with its documented behavior. The code is analyzed using LLMs to assess whether the implementation matches the expected behavior described in the docstring.
Related contents:
On-device LLM execution in React Native with Vercel AI SDK compatibility.
Run LLMs on-device in React Native via MLC LLM Engine with compatibility for Vercel AI SDK.
Artificial Neural Engine Machine Learning Library.
ANEMLL (pronounced like "animal") is an open-source project focused on accelerating the porting of Large Language Models (LLMs) to tensor processors, starting with the Apple Neural Engine (ANE).
The PM for your AI agent. An AI-powered task-management system you can drop into Cursor, Lovable, Windsurf, Roo, and others.
Taskmaster is your AI's personal project manager. Organize, research, expand, prioritize, and ship tasks effortlessly. Enjoy permanent context, zero drift & instant clarity. It's free, open-source & API-friendly. Let's vibe productively.
free AI coding. Non-agentic 100% free & open source coding tool for AI-assisted programming.
Copy folders and files for chatbots or initialize them hands-free! Use built-in API Tools for code completions, file refactoring and applying chat responses.
I made my AI think harder by making it argue with itself repeatedly. It works stupidly well.
CoRT makes AI models recursively think about their responses, generate alternatives, and pick the best one. It's like giving the AI the ability to doubt itself and try again... and again... and again.
Transparent AI, Rooted in Research, Open to All. The Open Source Deep Researcher Tool. AI-Powered Online Data Information Synthesis Assistant.
CleverBee is a powerful Python-based research assistant agent using Large Language Models (LLMs) like Claude and Gemini, Playwright for web browsing, and Chainlit for an interactive UI. It performs research assistance by browsing the web, extracting content (HTML), cleaning it, and synthesizing findings based on user research topics.
A self-contained, lightweight and OOB research platform for modern ML.
Boson is a lightweight, fully containerized, and feature-rich machine learning research platform. It centralizes essential tools to help teams keep projects lean, organized, and reproducible—while reducing overhead and boosting productivity. Think Databricks/Sagemaker but local and free.
Boson enables engineers and researchers to iterate faster without getting bogged down by infrastructure or tooling complexity.
Open-Source Browser Agent for autonomously performing complex tasks on the web
AI-powered multi-agent builder. Let AI build multi-agent systems for you Build with Agents SDK. Connect MCP tools. Open-source.
AI Autonomous Coding Agent for VS Code.
Autonomous coding agent right in your IDE, capable of creating/editing files, executing commands, using the browser, and more with your permission every step of the way.
Related contents:
Fully open-source command-line AI assistant inspired by OpenAI Codex, supporting local language models.
Open Codex is a fully open-source command-line AI assistant inspired by OpenAI Codex, supporting local language models like phi-4-mini.
Containerized versions of hundreds of MCP servers 📡 🧠
While experimenting with Model Context Protocol (MCP) servers, we found that setting them up could be a bit tedious and time-consuming. To simplify the process, we created containerized versions of these servers—making it quick, easy, and secure for anyone to get started.
Your agentic CLI developer.
Sidekick is an agentic CLI-based AI tool inspired by Claude Code, Copilot, Windsurf and Cursor. It's meant to be an open source alternative to these tools, providing a similar experience but with the flexibility of using different LLM providers while keeping the agentic workflow.
Exmeralda helps you ask questions about Elixir libraries and get accurate, version-specific answers. Powered by Retrieval-Augmented Generation (RAG), it combines the latest AI with real documentation to deliver helpful, grounded responses.
A 100-line minimalist LLM framework for Agents, Task Decomposition, RAG, etc.
Related contents:
Agents are AI systems that can perform workflows on users' behalf with a high degree of independence. This guide, designed for product and engineering teams, explores how to build agents with frameworks for identifying promise use cases, clear patterns for designing agent logic and orchestration, and best practices to ensure agents run safely, predictably, and effectively. It provides the foundational knowledge needed to confidently start building agents. Building reliable agents means starting with strong foundations - start small, validate with real users, and grow capabilities over time.
Principles for building reliable LLM applications.
What are the principles we can use to build LLM-powered software that is actually good enough to put in the hands of production customers?
Claude Code is Anthropic’s new agentic coding tool. A command-line companion that can read, edit, and manage your projects through natural language prompts. It integrates directly into your development environment, removing friction and helping you focus on what matters most: writing great code.
Related contents:
- Poking Around Claude Code @ Han, Not Solo.
- What makes Claude Code so damn good (and how to recreate that magic in your agent)!? @ minusx.
- Claude Code: Best practices for agentic coding @ Engineering at Anthropic.
- Agentic Coding Recommendations @ Armin Ronacher's Thoughts and Writings.
- Quand Claude Code pilote votre terminal... @ Korben :fr:.
- Turning Claude Code Into My Best Design Partner @ Between the Prompts.
- Using Claude Code to modernize a 25-year-old kernel driver @ Dmitry Brant.
- The babysitter problem @ chrisbeckman.dev.
- Claude Skills are awesome, maybe a bigger deal than MCP @ Simon Willison’s Weblog.
- Mastering Claude Code in 30 minutes @ Anthropic's YouTube.
Crafting Effective Prompts for Agentic AI Systems: Patterns and Practices.
🧠 Curated collection of system prompts for top AI tools. Perfect for AI agent builders and prompt engineers. Incuding: ChatGPT, Claude, Perplexity, Manus, Claude-Code, Loveable, v0, Grok, same new, windsurf, notion, and MetaAI.
A simple CLI to help you remember commands.
Zev helps you remember (or discover) terminal commands using natural language.
LocalAI is the free, Open Source OpenAI alternative. LocalAI act as a drop-in replacement REST API that’s compatible with OpenAI API specifications for local inferencing. It allows you to run LLMs, generate images, audio (and not only) locally or on-prem with consumer grade hardware, supporting multiple model families and architectures. Does not require GPU. It is created and maintained by Ettore Di Giacinto.
A lightweight next-gen data explorer - Postgres, MySQL, SQLite, MongoDB, Redis, MariaDB, Elastic Search, and Clickhouse with Chat interface
Related contents:
The first and the best multi-agent framework. Finding the Scaling Law of Agents
Building Multi-Agent Systems for Task Automation.
🐫 CAMEL is an open-source community dedicated to finding the scaling laws of agents. We believe that studying these agents on a large scale offers valuable insights into their behaviors, capabilities, and potential risks. To facilitate research in this field, we implement and support various types of agents, tasks, prompts, models, and simulated environments.
CAMEL emerges as the earliest LLM-based multi-agent framework, and is now a generic framework to build and use LLM-based agents for real-world task solving. We believe that studying these agents on a large scale offers valuable insights into their behaviors, capabilities, and potential risks. To facilitate research in this field, we implement and support various types of agents, tasks, prompts, models, and simulated environments.
MCP server for your browser.
Browser MCP is a Model Context Provider (MCP) server that allows AI applications to control your browser.
If you want to automate actions on a website, like repeatedly fill out a form, you normally can't do it with AI apps like Cursor or Claude because they don't have access to a web browser. With Browser MCP, you can connect AI apps to your browser so they can automate tasks on your behalf.
Turn any GitHub repository into a comprehensive AI-powered documentation hub.
Generate beautiful, world-class documentation from any GitHub repository — instantly. Just replace hub with summarize in any GitHub URL to generate a live, interactive documentation hub.
A CLI/API/WebUI tool designed to generate and autofill tables using AI in a scalable and diverse way.
An SDK for working with LLMs and AI Agents from Apache Airflow, based on Pydantic AI.
It allows users to call LLMs and orchestrate agent calls directly within their Airflow pipelines using decorator-based tasks. The SDK leverages the familiar Airflow @task syntax with extensions like @task.llm, @task.llm_branch, and @task.agent.
Transform Al Prototypes into Enterprise-Grade Products.
Langtrace is an Open Source Observability and Evaluations Platform for Al Agents.
Related contents:
LLMs for language and code + Time series and geospatial foundation models.
Achieve over 90% cost savings with Granite's smaller and open models, designed for developer efficiency. Fit for purpose and open sourced, these enterprise-ready models deliver exceptional performance against safety benchmarks and across a wide range of enterprise tasks from cybersecurity to RAG.
Related contents:
A Copilot for Git.
Cocommit is a command-line tool that works with your HEAD commit and leverages an LLM of your choice to enhance commit quality.
A good commit consists of multiple elements, but at a minimum, it should have a well-crafted commit message. Cocommit analyzes the message from the last (HEAD) commit and suggests improvements, highlighting both strengths and areas for enhancement.
Playwright Tools for MCP.
A Model Context Protocol (MCP) server that provides browser automation capabilities using Playwright. This server enables LLMs to interact with web pages through structured accessibility snapshots, bypassing the need for screenshots or visually-tuned models.
Related contents:
Balance agent control with agency. Build resilient language agents as graphs.
Gain control with LangGraph to design agents that reliably handle complex tasks. Build and scale agentic applications with LangGraph Platform.
LangGraph — used by Replit, Uber, LinkedIn, GitLab and more — is a low-level orchestration framework for building controllable agents. While langchain provides integrations and composable components to streamline LLM application development, the LangGraph library enables agent orchestration — offering customizable architectures, long-term memory, and human-in-the-loop to reliably handle complex tasks.
Related contents:
We are uncovering a new way of building software by embracing AI, iteration, and human intuition.
This is a cleanroom deobfuscation of the official Claude Code npm package.
Related contents:
Fleur is the app store for Claude. The easiest way to discover and install MCPs.
Fleur is a desktop application that serves as an app marketplace for MCPs. It allows you to discover, install, and manage apps that extend the functionality of Claude Desktop and Cursor.
All without having to use a command line. Fleur is made for non-technical users in mind, but is open-source and extensible so developers can make it their own.
Related contents:
MCP server for fetch web page content using Playwright headless browser.
A Datacenter Scale Distributed Inference Serving Framework.
NVIDIA Dynamo is a high-throughput low-latency inference framework designed for serving generative AI and reasoning models in multi-node distributed environments. Dynamo is designed to be inference engine agnostic (supports TRT-LLM, vLLM, SGLang or others) and captures LLM-specific capabilities.
Related contents:
AI Invisible Companion.
Amurex is your simple yet powerful AI meeting assistant that seamlessly integrates into your workflow. Built with cutting-edge AI, Amurex ensures you never miss a detail, always stay on top of action items, and make every meeting more productive.
Your all-knowing guide that unpacks every PDF into clear, actionable insights.
Auntie PDF is a web application that helps users extract information and insights from PDF documents. With a sassy, helpful personality, Auntie PDF makes understanding complex documents easier and more engaging.
A delightful Ruby way to work with AI through a unified interface to OpenAI, Anthropic, Google, and DeepSeek. A delightful Ruby way to work with AI. No configuration madness, no complex callbacks, no handler hell – just beautiful, expressive Ruby code.
Related contents:
A powerful AI-powered research assistant that performs deep, iterative analysis using multiple LLMs and web searches. The system can be run locally for privacy or configured to use cloud-based LLMs for enhanced capabilities.
Related contents:
AI-Powered Mobile App Builder.
Transform your ideas into polished mobile applications with our intuitive mobile app generator. No complex setup required.
A non-saturating, open-ended environment for evaluating LLMs in Factorio.
Large Language Models (LLMs) are rapidly saturating existing benchmarks, necessitating new open-ended evaluations. We introduce the Factorio Learning Environment (FLE), based on the game of Factorio, that tests agents in long-term planning, program synthesis, and resource optimization.
A native macOS app that allows users to chat with a local LLM that can respond with information from files, folders and websites on your Mac without installing any other software. Powered by llama.cpp.
Chat with an local LLM that can respond with information from your files, folders and websites on your Mac without installing any other software. All conversations happen offline, and your data is saved locally.
Affiche une alerte sur des sites générés tout ou partie par des IA. Liste gérée par Next.ink.
Related contents:
Evolving agents is a production-grade environment for orchestrating, evolving, and managing AI agents.
A production-grade framework for creating, managing, and evolving AI agents with intelligent agent-to-agent communication. The framework enables you to build collaborative agent ecosystems that can semantically understand requirements, evolve based on past experiences, and communicate effectively to solve complex tasks.
An in-depth book and reference on building agentic systems like Claude Code. A deep-dive guide into architecture patterns for building responsive, reliable AI coding agents.
There's been a lot of asking about how Claude Code works under the hood. Usually, people see the prompts, but they don't see how it all comes together. This is that book. All of the systems, tools, and commands that go into building one of these.
A practical deep dive and code review into how to build a self-driving coding agent, execution engine, tools and commands. Rather than the prompts and AI engineering, this is the systems and design decisions that go into making agents that are real-time, self-corrective, and useful for productive work.
Terminal-based AI coding tool that can use any model that supports the OpenAI-style API.
The Platform for Building Stateful Agents. Build agents with infinite context and human-like memory, that can learn from data and improve with experience. Letta (formerly MemGPT) is a framework for creating LLM services with memory.
👾 Letta is an open source framework for building stateful LLM applications. You can use Letta to build stateful agents with advanced reasoning capabilities and transparent long-term memory. The Letta framework is white box and model-agnostic.
Related contents:
Finding the Scaling Laws of Agents. The first and the best multi-agent framework.
🐫 CAMEL is an open-source community dedicated to finding the scaling laws of agents. We believe that studying these agents on a large scale offers valuable insights into their behaviors, capabilities, and potential risks. To facilitate research in this field, we implement and support various types of agents, tasks, prompts, models, and simulated environments.
The framework enables multi-agent systems to continuously evolve by generating data and interacting with environments. This evolution can be driven by reinforcement learning with verifiable rewards or supervised learning.
Ollama Automated Security Intelligence Scanner.
🛡️ An AI-powered security auditing tool that leverages Ollama models to detect and analyze potential security vulnerabilities in your code.
Advanced code security analysis through the power of AI
Related contents:
Go beyond nascent AI demos. The intelligent AI-native gateway for prompts and agentic apps.
Effortlessly build AI apps that can answer questions and help users get things done. Arch is the AI-native proxy that handles the pesky heavy-lifting so that you can move faster in building agentic apps, prevent harmful outcomes, and rapidly incorporate latest models.
AI-native (edge and LLM) proxy for agents. Move faster by letting Arch handle the pesky heavy lifting in building agentic apps -- ⚡️ query understanding and routing, seamless integration of prompts with tools, and unified access and observability of LLMs. Built by the contributors of Envoy proxy.
Prompt, run, edit, and deploy full-stack web applications using any LLM you want!
Nut is an open source fork of Bolt.new for helping you develop full stack apps using AI. AI developers frequently struggle with fixing even simple bugs when they don't know the cause, and get stuck making ineffective changes over and over. We want to crack these tough nuts, so to speak, so you can get back to building.