headroomlabs-ai/headroom
Headroom is a context compression layer for AI agents, reducing token usage by 60-95% across tool outputs, logs, RAG chunks, files, and conversation history, while maintaining accuracy.
Awesome AI Agents › End-to-End Optimizers
turing-machines/mentals-ai
Mentals AI is an innovative tool designed to create and operate AI agents using simple markdown files with a .gen extension, eliminating the need for traditional programming languages. The core concept revolves around defining agents as executable files where users focus solely on the agent's logic without writing scaffolding code in Python or other languages. This approach redefines foundational frameworks for future AI applications by enabling recursive loops, memory management, and tool integration within agents. The system supports complex multi-agent interactions, game generation, content creation, and more, showcasing its versatility. The project emphasizes three main differentiators: the Agent Executor operates through a recursive loop where the language model decides the next steps; agents can be created entirely in markdown, with optional Python integration; and it serves as a blank canvas for creating and integrating various reasoning frameworks like Tree of Thoughts, ReAct, Self-Discovery, and Auto-CoT. This flexibility allows users to build custom reasoning frameworks or combine existing ones into complex sequences. Key concepts include instructions (prompts) that form the basic components of agents, working memory (context) that maintains state across instruction calls, and native tools for message output, user input, file handling, Python execution, Bash commands, and short-term memory. The project also features a local vector database for storing chats and private information, and a web UI is in development. Mentals AI supports Llama3 and OpenAI API models, requiring an OpenAI API key for operation. It is designed to be cross-platform with dependencies like libcurl, libfmt, pgvector, and poppler. The project is open-source and actively developed, aiming to provide a flexible, powerful platform for AI agent creation and operation through a unique markdown-based approach.
https://github.com/turing-machines/mentals-ai
Also in End-to-End Optimizers
topoteretes/cognee
Cognee is an open-source project that provides scalable and modular memory solutions for AI agents using ECL pipelines, enabling efficient data ingestion, knowledge graph generation, and query capabilities to enhance AI performance and reduce hallucinations.
letta-ai/letta
Letta is an open-source framework for building stateful AI agents with advanced reasoning, transparent long-term memory, and context management, supporting multiple LLM backends and providing a graphical development environment.
langchain-ai/opengpts
OpenGPTs is an open-source platform that replicates and extends OpenAI's GPTs and Assistants API, offering customizable cognitive architectures, extensive tooling, and support for multiple language models and vector databases.
aiwaves-cn/agents
Agents 2.0 is an open-source framework that enables the training and self-evolution of autonomous language agents through symbolic learning inspired by neural network methodologies.
getzep/zep
Zep is an AI memory foundation that uses a temporal Knowledge Graph to enable continuous learning and personalized experiences for AI agents by efficiently managing and retrieving contextual chat and data artifacts.
algorithmicsuperintelligence/optillm
OptiLLM is an optimizing inference proxy that significantly improves the accuracy and performance of large language models on reasoning tasks without requiring any model training, by applying over 20 advanced optimization techniques during inference.
mll-lab-nu/RAGEN
A reinforcement learning framework for training multi-turn reasoning LLM agents on the StarPO algorithm, shipping ten Gym-compatible environments plus diagnostics for echo-trap and reasoning-collapse failure modes.