Top AI Open Source Projects to Watch in 2025

A powerful Model Context Protocol (MCP) command-line interface tool

Python library providing a memory layer for AI applications, supporting MongoDB integration and semantic search functionality.

Cursor and Windsurf community rules and MCP server directory platform

A multi-agent financial trading framework based on large language models, simulating the collaborative structure of real trading firms.

A complete tutorial on building AI agents from scratch, including email assistants, human-AI collaboration, memory functions, and Gmail integration.

A deep learning project that generates realistic human avatars from audio, capable of synthesizing full-body gestures and facial expressions from speech.

High-performance Node.js/TypeScript web crawler optimized for LLMs, supporting multi-engine crawling and structured data extraction.

Netflix-level video translation, localization, and dubbing tool, realizing AI subtitle cutting, translation, alignment, and dubbing with one click.

A blazing fast AI search engine built on the Firecrawl web crawling API, providing intelligent answers with real-time citations.

A natural language-based intelligent search tool for React codebases, supporting code finding and analysis for JS/TS projects.

AI terminal pair programming assistant that lets you collaborate with large language models to write and modify code

An AI programming assistant that automatically converts GitHub issues and feature requests into code changes and pull requests.

A natural language-based AI code generation command-line tool that allows AI to automatically write and execute code.

A lightweight AI agent framework that automates intelligent tasks through Python code execution.

FLUX.1 is an advanced text-to-image generation AI model developed by Black Forest Labs, supporting high-quality image generation and editing.

MCP server providing Figma layout information for AI programming assistants, allowing tools like Cursor to directly access Figma design data.

A powerful tool designed specifically for creating fine-tuning datasets for large language models, supporting intelligent document processing, question generation, and multi-format export.

AI-powered real-time face swap and one-click video deepfake tool, achievable with just a single image

An autonomous research agent powered by large language models, capable of conducting in-depth local and web research on any topic and generating detailed reports with citations.

A lightweight GPT training framework that allows you to train a small 26M parameter language model from scratch in just 2 hours.

Open-source AI-powered intelligent search engine, a free alternative to Perplexity AI

LLM API management and distribution system, supporting unified interface management for various large language models

Microsoft's large-scale self-supervised pre-trained unified language model, supporting foundational model research across tasks, languages, and modalities.

An open-source personal AI assistant that can search documents and the web, supports multiple LLM models, and has intelligent search, dialogue, and automation capabilities.

An AI-powered, open-source, end-to-end web application testing framework that uses visual AI agents to automate the testing process.