An open-source, code-first Python toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control.

Open Source DeepWiki: AI-Powered Wiki Generator for GitHub/Gitlab/Bitbucket Repositories. Join the discord: https://discord.gg/gMwThUMeme

Awesome curated collection of images and prompts generated by GPT-4o and gpt-image-1. Explore AI generated visuals created with ChatGPT and Sora, showcasing OpenAI’s advanced image generation capabilities.

Open Source TypeScript AI Agent Framework

OmniSVG is the first family of end-to-end multimodal SVG generators that leverage pre-trained Vision-Language Models (VLMs), capable of generating complex and detailed SVGs, from simple icons to intricate anime characters.

Image editing is worth a single LoRA! 0.1% training data for fantastic image editing! Training released! Surpasses GPT-4o in ID persistence! Official ComfyUI workflow release! Only 4GB VRAM is enough to run!

A SOTA open-source image editing model, which aims to provide comparable performance against the closed-source models like GPT-4o and Gemini 2 Flash.

LAYRA—an enterprise-ready, out-of-the-box solution—unlocks next-generation intelligent systems powered by visual RAG and limitless visual multi-step agent workflow orchestration.

Nexent 是一个可以零代码开发智能体的平台 —— 无需编排，无需复杂拖拉拽。基于 MCP 工具生态，Nexent 还提供强大的模型集成、数据处理、知识库管理能力。

An iterative multi-LLM consensus framework for accurate cell type annotation in single-cell RNA-seq data

Atropos is a Language Model Reinforcement Learning Environments framework for collecting and evaluating LLM trajectories through diverse environments

🐸 Read Frog - Open Source Immersive Translate | 🐸 陪读蛙 - 开源沉浸式翻译

Implementing DeepSeek R1's GRPO algorithm from scratch

Live speech translation application built with Electron 34 and React, using OpenAI's Realtime API.

Proxy remote LLM API as Ollama and LM Studio, for using them in JetBrains AI Assistant

Voice AI agent starter kit with Groq, Llama 4, and (optionally) Twilio

Official PyTorch implementation of the paper "Chapter-Llama: Efficient Chaptering in Hour-Long Videos with LLMs"

Summarize PDFs into beautiful sections with Llama 3.3

eShopLite is a set of reference .NET applications implementing an eCommerce site with features like Semantic Search, MCP, Reasoning models and more.

Helm charts for llm-d

A cutting-edge voice assistant combining real-time speech recognition, AI reasoning, and neural text-to-speech capabilities.

🔥 Official implementation of "Generate, but Verify: Reducing Visual Hallucination in Vision-Language Models with Retrospective Resampling"

VelociraptorMCP is a Model Context Protocol bridge for exposing LLMs to MCP clients.

[Support 0.48.x]（Reset Cursor AI MachineID & Auto Sign Up / In & Bypass Higher Token Limit）自动注册 Cursor Ai ，自动重置机器ID ， 免费升级使用Pro功能: You've reached your trial request limit. / Too many free trial accounts used on this machine. Please upgrade to pro. We have this limit in place to prevent abuse. Please let us know if you believe this is a mistake.

NoteCraft is a full-stack web app built with Django, Celery, and Next.js that lets users upload academic PDFs, generate AI-powered notes, and retrieve relevant content using RAG. It's optimized for long documents, supports async processing, and runs fully containerized with Docker.

✨ PyTorch implementation of "Cora: Correspondence-aware Image Editing Using Few-Step Diffusion", accepted at SIGGRAPH 2025.

A Model Context Protocol (MCP) server that enables AI assistants to generate images, text, and audio through the Pollinations APIs. Supports customizable parameters, image saving, and multiple model options.

Enable tool-use ability for any LLM model (DeepSeek V3/R1, etc.)

First-of-its-kind benchmark for evaluating the protection capabilities of large language model (LLM) guard systems

Node.js+Playwright服务器，通过模拟 OpenAI API 的方式来访问 Google AI Studio 网页版，服务器无缝交互转发gemini模型对话。这使得兼容 OpenAI API 的客户端（如 Open WebUI, NextChat 等）可以使用 AI Studio 的无限额度及能力。经测试因无法绕过自动化检测故暂不支持无头模式启动实例-自用项目随缘维护

Used for thinking process intervention of reasoning models such as DeepSeek-R1, effectively controlling the reasoning thinking process. 用于DeepSeek-R1等推理模型的思维过程干预，有效控制推理思考过程

Swift local LLM client for iOS, macOS, Linux

An external provider for Llama Stack allowing for the use of RamaLama for inference.

A powerful Android app that connects to LM Studio running on your computer, allowing you to chat with your favorite AI models from your mobile device.

SPICE: A Synergistic, Precise, Iterative, and Customizable Image Editing Workflow

awsome kali MCPServers is a set of MCP servers tailored for Kali Linux, designed to empower AI Agents in reverse engineering and security testing. It offers flexible network analysis, target sniffing, traffic analysis, binary understanding, and automation, enhancing AI-driven workflows.

[AAAI2025] CAD-GPT: Synthesising CAD Construction Sequence with Spatial Reasoning-Enhanced Multimodal LLMs

Gen AI-Powered Virtual Try-On Clothes Platform Upload any model and garment image to preview realistic try-on results instantly. Built with Google Gemini, FastAPI, and React. Ideal for fashion, retail, and e-commerce.

This repository hosts the instructions and workshop materials for Lab 331 (Deep Research with Langchain and DeepSeek R1) for Microsoft Build

Agentic Coding Rules, Templates etc...

OmniMind: An open-source Python library for effortless MCP (Model Context Protocol) integration, AI Agents, AI workflows, and AI Automations. Plug & Play AI Tools for MCP Servers and Clients, powered by Google Gemini.

The home of HeadlineSquare and Dr. Headline. HeadlineSquare is a public square for US news headlines, fully powered by Dr. Headline, an autonomous AI agent who applies academic neutrality and rigor to news curation.

A simple, user-friendly CLI tool for managing, and running, llama.cpp's llama-server with multiple configuration profiles and providing OpenAI-compatible API access.

Faraday: An Autonomous Web Research Agent (LangGraph/Streamlit). 🕵️♀️ Investigates queries using dynamic tools (Tavily, Google, NewsAPI, etc.), gathers multi-source info, and synthesizes structured reports in a Streamlit UI. Features agentic workflow & source tracking.

🌟 Full-stack app for real-time avatar streaming with HeyGen & Gemini AI. Built with React, TypeScript, Express, and Tailwind during a hackathon by Team "Bit by Bit"

SunoBot — a personalized, generative audio companion designed to boost user engagement through interactive storytelling and voice-based experiences.

A cutting-edge Cascading voice assistant combining real-time speech recognition, AI reasoning, and neural text-to-speech capabilities.

在Mac上记录您在任何应用打下的字与您剪切板的内容并在本地通过ollama DeepSeek-r1分析。Reminding all the words you have typed for Macbook and analysis with ollama Deepseek-r1

Prompt Booster: A comprehensive tool for optimizing LLM prompts with version control, A/B testing, and template management. Supports multiple AI providers (OpenAI, Gemini, DeepSeek, Qwen, etc.) across web and desktop platforms. Increase your AI prompt effectiveness with professional engineering tools.

This project enables real-time streaming of audio (and optionally video or screen captures) from your local device to Google Gemini using the Live API. It allows you to interact with Gemini through both text and voice, supporting conversational AI responses.

A chat pdf app on MacOS, keeping your data safe, handling massive documents, and connecting you with SOTA AI models

An AI-Powered Multi-Agent Solution for Retail Banking Products.

code editor with completion powered by llamacpp

use florence model for few shot special logo detection; iou reward training with grpo for florence and VLMs

an AI-powered desktop application designed to help users discover and optimize system prompts for Large Language Models (LLMs) without requiring complex fine-tuning or LoRA training.

"Open-source toolkit (Python Library, Registry API, CLI) for secure, decentralized AI agent interoperability using A2A/MCP."

A comprehensive security checklist for MCP-based AI tools. Built by SlowMist to safeguard LLM plugin ecosystems.

llama.cpp server in a Python wheel.

A RAG video chatbot to deliver accurate, context-aware answers. Users can upload videos, which are split into chunks and embedded into a vector database. When a question is asked, the chatbot retrieves relevant chunks using similarity search and generates response.

⽊ ongoing list of curated resources for decentralized AI and LLMs, with a focus on agents, blockchains, and privacy 🏴‍☠️

Learning Management System for CSI-300: Database Management Systems

Fire-Detection-Siglip2 is an image classification vision-language encoder model fine-tuned from google/siglip2-base-patch16-224 for a single-label classification task. It is designed to detect fire, smoke, or normal conditions using the SiglipForImageClassification architecture.

Cook your files into AI-ready servings

基于DeepSeek-R1黑盒蒸馏的网络安全渗透领域推理模型。可高效的应对断网情况下的网络安全大赛。简介写完整了，图片加载不出来看看是否梯子挂好了

Inference Llama 2 in several files of pure C but on a C64

A privacy-first, self-hosted AI chat UI powered by Ollama and open-source models like Mistral, LLaMA 3, and Phi. Designed to be fast, lightweight, and simple to deploy (Docker-ready)—no GPU or complex setup required. Spin up your own ChatGPT alternative in minutes with a beautiful TailwindCSS UI and Go Fiber backend. Supports real-time streaming

A flexible PHP library providing access to various AI models (Gemini, OpenAI, Anthropic, DeepSeek, Grok) via a consistent interface.

This project demonstrates a Generative AI-powered assistant that streamlines the job application process using Google Gemini Pro. It analyzes a user’s resume against a job description, calculates a match score, suggests tailored bullet points, and generates a personalized cover letter — all formatted in structured JSON for automation.

YouTube AI Timecodes Generator

a lua runtime for vibecoding

​​LibRAG​​: A next-gen intelligent content retrieval engine that redefines traditional RAG architecture. Leveraging ​​full-chain LLM inference​​ to streamline document processing and paragraph recall, it eliminates embedding models/vector databases while achieving precise end-to-end semantic retrieval through deep language understanding .

一个福寿螺卵自动统计系统，用户只需上传一组或更多带有福寿螺卵的图片即可统计出每张图片里的福寿螺卵数量。An automatic statistics system for Fushou snail eggs. Users only need to upload a set or more pictures with Fushou snail eggs to count the number of Golden Apple Snail eggs in each picture.

This is a Python program that automatically generates an "awesome list" for a specific keyword as a markdown file. An "awesome list" is a list of resources related to a specific topic. Currently, the resources include GitHub projects, Google Scholar articles, YouTube videos, courses, slides and presentations, software and tools and podcasts. The aw

Here is where I store all my useful prompts

Un Asistente de IA basado en Google Gemini para Android Antiguo 4.2-8.1 / An AI Assistant based on Google Gemini for Old Android 4.2-8.1

CLI AI "agents"™ + Unix philosophy 🐙

A production-ready FastAPI template for building AI agent applications with LangGraph integration. This template provides a robust foundation for building scalable, secure, and maintainable AI agent services.

AI-driven academic portfolio with automated content aggregation and intelligent web mentions discovery - like Yahoo Pipes for academics, but with GPT-4 superpowers! A modern Jekyll site that mashes up publication APIs, social feeds, and scholarly references into one sleek, auto-updating research showcase.

新版GPT-4o生图功能太强了（附国内使用指南）【6月最新更新】

Simplified "how to run" Llama 4

DeepSeek | 中文官网、DeepSeek网页版、API 调用和本地部署教程 | 最全使用指南~【2025年4月更新】

A conversational AI agent built with Python, Streamlit, and LangChain, featuring multi-LLM support (Gemini, Llama 3), persistent chat history via PostgreSQL, agent tools (Web Search, Python REPL), and image generation using Google Cloud's Imagen model.

Ai Vr Machine Learning Natural language Poker Coach

An intelligent web application that transforms complex questions into interactive visual mind maps using AI (Deepseek integration).

DeepSeek-R1 免费的API Key可使用。https://hao666666.online 获取免费API接口。

no-llm is a lightweight library designed to simplify the integration of machine learning models without relying on large language models. It provides essential tools for developers to create efficient, scalable applications while maintaining clear and concise code.

An interactive Streamlit app that lets users upload PDFs and ask questions to receive concise, reasoning-based answers using the DeepSeek-R1 language model.

A fine-tuned version of SmolLM2-360M-Instruct-bnb-4bit specialized for parsing unstructured calendar event requests into structured JSON data.

A fast, lightweight CLI tool for versioning, testing, and optimizing your AI prompts across multiple providers. Easily track prompt evolution, run A/B tests, and measure performance without Git dependencies. Supports OpenAI, Claude, Llama, and HuggingFace.

Fluent Go prompt builder for LLaMA- and Chat-style models with role-delimited blocks and built-in list formatting.

A Python-based application that enables debates and answer optimization between two AI language models. The platform supports OpenAI models, Anthropic Claude models, and various Chinese AI models like DeepSeek, Moonshot, ChatGLM, Qwen, and ERNIE.