Open-source repositories created in the last 90 days, ranked by the traction they're picking up. Refreshed nightly from GitHub.
This repo provides a flexible framework where AI agents can easily integrate various tools and functionalities through a plugin system, making them highly adaptable.
This repo provides a plugin that gives AI agents free, unlimited access to advanced language models without any API keys or logins.
This repo uses AI agents to automate the entire video content creation and adaptation workflow, from concept to distribution.
It acts as a unified interface for integrating and leveraging multiple LLM models from different providers, simplifying access for users and developers.
This repo provides a specialized headless browser designed for AI agents to interact with websites efficiently and reliably.
This repo uses AI agents to translate natural language into robust Android automation, enabling complex mobile workflows.
This repository showcases highly optimized, resource-efficient inference for a massive machine learning model on basic hardware.
It's an AI-powered framework for automated content curation and generation, specifically for news aggregation and trend analysis, leveraging LLMs for summarization.
This 'world model' likely utilizes RAG principles to retrieve and synthesize vast amounts of enterprise data, forming a comprehensive and dynamic understanding of complex systems.
It demonstrates an intelligent LLM application for real-time communication assistance, focusing on intent analysis, risk assessment, and context-aware reply generation.
This repo provides foundational infrastructure for building AI agents that can continuously learn and enhance their capabilities over time.
This repo combines ChatGPT's planning intelligence with Codex's execution capabilities to create a more effective AI agent for programming tasks.
This repo provides a set of rules for AI agents to improve the quality of AI-generated content by filtering out generic or low-quality outputs.
It provides a highly optimized runtime for specific machine learning decision models, emphasizing local and rapid inference.
It provides a platform for creating autonomous AI 'coworkers' that are equipped with their own virtual computers and tools, allowing them to perform tasks independently.
This suite provides advanced routing and reasoning capabilities for AI agents, allowing them to intelligently select and apply the right tools and models for specific tasks.
It integrates AI agents directly into common productivity applications, making smart assistance accessible for everyday office tasks.
This repo provides an invaluable open-source resource for understanding the intricate quantitative design of large language model (LLM) infrastructure, bridging the gap between theoretical models and practical hardware implementation.
This repository explicitly features RAG as a core component of its comprehensive learning path for large language models, indicating its relevance for practical LLM development.
It is a curated gallery of visual summaries from leading machine learning research papers, serving as a resource for researchers in the field.
It provides a way to integrate and run a specific type of machine learning decision model within web development environments.
It offers an open-source, highly efficient machine learning model specifically designed for very fast, local decision-making.
The notebooks provide hands-on examples and practical implementations of RAG, directly addressing a key skill for AI engineers.
It offers a specialized, efficient machine learning model tailored for quick, confident decision-making within AI agent systems.
It is an educational resource focused on teaching fundamental concepts and principles across various domains of machine learning.
It serves as a comprehensive resource and curated guide for a specific and prominent machine learning decision model, including tools and evaluations.
The project focuses on distilling and retrieving personal knowledge and memories for an AI agent, which is a direct application of RAG principles for personalized AI.
It's a distributed system that could leverage machine learning for 'intent routing' and optimization within complex automated workflows.
The 'knowledge learning' module within this AI workstation suggests a strong reliance on retrieving and processing information, aligning with RAG for enhanced understanding and creativity.
The platform's goal of 'solidifying' professional experience and processes into AI digital employees heavily depends on ingesting, retrieving, and applying domain-specific knowledge, which is a core RAG function.