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
James4Ever0/agi_computer_control
Cybergod is an ambitious project aimed at creating the first autonomous computer program capable of performing any task to earn money without human intervention. It is designed as an Artificial General Intelligence (AGI) system that can seamlessly integrate into human society by replacing half of the human population's functions without bias based on income, rank, or education. The project emphasizes a multi-phase training approach: starting with agent pre-training where the agent acts randomly in virtual and physical spaces, followed by post-training to align with human preferences, reinforcement learning through games and challenges, a General Turing test phase where the agent must earn real-world currency, and finally a Free AI phase that tests the agent against essential freedoms. Cybergod features advanced terminal interaction capabilities built on top of tmux and other libraries, allowing it to understand and manipulate terminal interfaces in both text-only and multimodal modes. It supports precise cursor placement even with multi-width Unicode characters and can render cursors correctly in GUI environments. The system collects detailed input/output statistics from terminal sessions, enabling the development of intelligent event-driven terminal agents. The project also integrates an economy system powered by a digital currency called Godcoin and a smart contract system that allows both humans and agents to publish contracts and delegate tasks. The project is supported by extensive documentation, demos, and tools for terminal environment integration, including Python libraries for interacting with terminal sessions programmatically. It aims to provide a more frequent and practical usage experience than popular AI chatbots like ChatGPT. Cybergod is also backed by academic publications and a dataset called The Frozen Forest used for training. Overall, it represents a comprehensive effort to build a fully autonomous, economically active AGI agent capable of operating independently in real-world environments.
https://github.com/James4Ever0/agi_computer_control
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.