#llm Repositories
2,314 repositories total
Rent the model. Own everything else. Free, open-source AI app for real work.
Building a MiniLLM from Scratch (Pretraining + SFT + DPO in Practice)
Memory Bridge MemoryBridge โ A cross-device, cross-platform shared memory layer for AI | Cross-device and cross-platform shared memory for AI (CDSMP implementation)
This repository contains code to quantitatively evaluate instruction-tuned models such as Alpaca and Flan-T5 on held-out tasks.
A skill for AI agents: search the web with SearXNG, browse with Camofox, bypass protections with CloakBrowser. Anti-hallucination by design. Self-hosted, free, unlimited.
๐ An intelligent e-commerce data warehouse AI Agent for natural-language queriesโthe ideal hands-on project for learning LangGraph systematically. Built with LangGraph, FastAPI, Qdrant, Elasticsearch, MySQL, and React, it delivers a complete implementation of a metadata knowledge base, hybrid retrieval, natural-language-to-SQL (NL2SQL) generation and validation, SQL execution, and streaming query results. Full-stack frontend and backend code that runs out of the box, one-click deployment in a Docker environment, plus the free ai-agents-from-zero tutorial and code branches for each chapter. Ideal for systematically learning LLM applications, data analytics agents, and enterprise-grade AI engineering in practice.
Plugin-first long-term memory for agent platforms โ reference implementation of AMP (Agent Memory Protocol).
๐ฅ๐ฅ๐ฅ A curated list of papers on LLMs-based multimodal generation (image, video, 3D and audio).
Summarize and query from a lot of heterogeneous documents. Any LLM provider, any filetype, advanced RAG, advanced summaries, scriptable, etc
An open-source enhanced version of Codex/Claude Code/Cursor, focused on completing complex, long-running tasksโfrom education, programming, and office work to daily life, entertainment, and gamingโwith a single sentence. Deploy it on your own server, and your team can code by opening a browserโeven on mobile. Dual entry points via CLI and Web UI, Multi-Agent collaboration, and native direct connections to domestic Chinese large language models such as Qwen and DeepSeek. Skills/plugins/cross-session memory, an eight-layer security sandbox, and data that never leaves your machines. One-click Docker deployment, MIT-licensed open source, and zero lock-in.
Local inference for diffusion language models (LLaDA, Dream), written in Mojo
Open-source whiteboard AI tutor with animated lessons and interactive playback. Compare Claude, Gemini & DeepSeek across 288 generation runs and 64 blind-reviewed boards.
Where terminal agents work together. Claude Code, Codex, Copilot CLI, and pi on the same task, in one mission, each keeping its own TUI in a real terminal.
A curated catalog of 230+ semantic anchors: well-defined terms like arc42, TDD London School, or tracer bullet that activate deep, consistent knowledge in LLMs. Use them in prompts, skills, and agent instructions instead of spelling out methods yourself.
ScreenAgent: A Computer Control Agent Driven by Visual Language Large Model (IJCAI-24)
A conversational, AI device + software framework for companionship, entertainment, education, healthcare, IoT applications, and DIY robotics. Built with Python, NextJS, Arduino, ESP32, LLMs (GPT-4o), Deepgram STT and Azure TTS ๐ค
An open collection of methodologies to help with successful training of large language models.
Open-source, self-hostable alternative to ChatGPT and NotebookLM โ agents, skills, MCP, and cited knowledge in one AI workstation. Your models, your infrastructure. Includes public MCP server and agent skills directories.
Chat floating-window assistant (macOS): screen awareness + a local small model to assess intent and risk, generating reply suggestions based on conversation scripts. Read-only.
An MCP for WireShark (tshark). Empower LLM's with realtime network traffic analysis capability
AWS Generative AI CDK Constructs are sample implementations of AWS CDK for common generative AI patterns.
Meet Mitsu, your chibi red-team operator. Connect your AI provider, set the scope, and she can take an authorized pentest from recon and attack-surface mapping through vulnerability discovery, > exploitation, and post-exploitation, armored with 165+ MCP-connected tools and live session intel. Guide every step or let her pwn autonomously. ( โ โฟโ )
GraphRAG-rs is a high-performance, state-of-the-art Rust implementation of GraphRAG (Graph-based Retrieval Augmented Generation) that builds knowledge graphs from documents and enables natural language querying with configurable entity extraction and local LLM integration
A curated guide to Generative Engine Optimization (GEO) resources: guides, tools & research to boost visibility in AI-powered search engines.
The open-source agent system that writes research papers with verified citations. Every citation is checked against Crossref, OpenAlex, and Semantic Scholar.
๐ท๏ธ The fastest open-source AI Gateway and LLM Gateway โ one binary with a built-in console to route, secure, budget, and observe every LLM call across OpenAI, Anthropic, Gemini, Bedrock, and more.
AI Thesis System ยท AI thesis generation and writing, one-click thesis creation, AI-generated content reduction, plagiarism reduction, thesis defense PPT ยท OEM white-label partnerships and franchising
Collect and curate a reliable list of Claude and OpenAI relay API resources to help developers quickly find stable, dependable, and cost-effective LLM API services.
Architect-centric agentic swarm plugin for OpenCode. Hub-and-spoke orchestration with SME consultation, code generation, and QA review.
Paper-Agent is an intelligent academic paper search and research tool for researchers and students. Built on a multi-agent collaboration architecture (LangGraph), it uses natural language processing (NLP) and automated search to help users efficiently find academic papers, analyze their content, and conduct literature reviews. Paper-Agent supports multi-platform integration, keyword search, automated analysis, and literature research, improving the efficiency of academic research. It is ideal for a wide range of applications, including paper writing, academic research, and research project management, making it an ideal assistant for academic research.
Self-hosted Enterprise and Personal AI + agent runtime in .NET (NativeAOT-friendly)
Turns the documentation you already write into tests that run. A failing test means your product and your docs disagree, and names the section.
A minimal yet powerful framework for creating AI agents with full control over tools, providers, and execution flow.
Client for OpenClaw โ Connect ClawWork to your own OpenClaw and unlock 10x multi-session productivity.
Deprecated historical repo. Superlinked now develops SIE, a self-hosted inference engine for embeddings, reranking, OCR, extraction, and document processing.
Official DESIGN.md files published by companies themselves. First-party design systems that AI agents read to generate consistent UI.