llm-internals
Learn LLM internals step by step - from tokenization to attention to inference optimization.
Un agent IA est un programme qui utilise un modèle de langage pour décider de l’étape suivante, appelle des outils comme un navigateur, un interpréteur de code ou une API, et recommence jusqu’à ce que la tâche soit terminée. Les projets présentés ici vont des frameworks pour créer vos propres agents aux agents prêts à l’emploi qui écrivent du code, naviguent sur le web ou exécutent des processus métier.
Comparez trois points : les modèles utilisables (un seul fournisseur, ou tout point de terminaison compatible OpenAI, y compris local), la manière de limiter ce que l’agent peut faire (bac à sable, étapes d’approbation, listes d’autorisations), et sa capacité à conserver une mémoire entre les étapes et les sessions. Un agent qui exécute du code ou des commandes shell doit tourner dans un bac à sable : vérifiez-le avant de lui donner accès à votre machine.
Learn LLM internals step by step - from tokenization to attention to inference optimization.
AIDE: an LLM agent for machine learning engineering - the research Weco grew out of. Referenced in OpenAI MLE-bench.
Open-source agent platform for Global × China enterprises — wire every system through one agent core. Self-hosted, any LLM.
VS Code extension for LLM-assisted code/text completion
A beautiful, cross-platform downloader for YouTube, TikTok, Instagram, and 1800+ sites (yt-dlp GUI) with AI video summaries and post-processing
Burp Suite extension that adds built-in MCP tooling, AI-assisted analysis, privacy controls, passive and active scanning and more
✨ AI-powered markdown editor - leverage LLMs with your documents - 100% local or in the cloud
Collect, organize, use, and share, all in OmniBox.
One sentence creates an AI-driven world — generate maps, characters, and watch stories emerge on their own. 一句话生成一个AI自主驱动的世界.
E2B Desktop Sandbox for LLMs. E2B Sandbox with desktop graphical environment that you can connect to any LLM for secure computer use.
Token-efficient MCP adapter for Pi coding agent
Local AI app and inference engine for agents. Run open-weight LLMs locally — private, 100% offline on your computer. Join our Discord: https://discord.com/invite/8wGSsvmg4V
The open document intelligence platform for builders and hackers - DMS for the agentic world
An AI co-worker with its own computer. Self-evolving, persistent memory, MCP server, secure credential collection, email identity. Built on the Claude Agent SDK.
Run local LLMs like llama, deepseek-distill, kokoro and more inside your browser
From Automated Idea Factory to Realization
Open-source observability tool that uses AI agents to self-heal your software
LLM powered fuzzing via OSS-Fuzz.
Samurai-inspired multi-agent system for Claude Code. Orchestrate parallel AI tasks via tmux with shogun → karo → ashigaru hierarchy.
🍦 Speech-AI-Forge is a project developed around TTS generation model, implementing an API Server and a Gradio-based WebUI.
BrowserWing turns your browser actions into MCP commands Or Claude Skill, allowing AI agents to control browsers efficiently and reliably. Say goodbye to slow, token-heavy LLM interactions — let agents call commands directly for faster automation. Perfect for AI-driven tasks, browser automation, and boosting productivity.
An open-source, code-first Typescript toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control.
Trace-native CI/CD for AI agents — production failures become regression tests that block the PR. Auto-detect, cluster, freeze into hermetic cases, replay in CI for $0.
A Model Context Protocol (MCP) server that enables secure interaction with MySQL databases