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
agiresearch/A-mem
A-MEM: Agentic Memory for LLM Agents is an innovative memory system designed to enhance the capabilities of Large Language Model (LLM) agents by providing a dynamic and sophisticated way to organize and utilize memories. Unlike traditional memory systems that offer basic storage and retrieval, this system introduces an agentic approach that allows LLM agents to manage memories dynamically and interact flexibly with their memory components. The system is inspired by the Zettelkasten method, enabling dynamic memory organization, intelligent indexing, and linking of memories through ChromaDB, a vector database for efficient semantic search. The project features comprehensive note generation with structured attributes, interconnected knowledge networks, and continuous memory evolution and refinement. It supports agent-driven decision-making for adaptive memory management, allowing the system to automatically analyze content relationships, update metadata and context, and create semantic connections between memories. This results in a more organized and contextually rich memory system that improves the agent's ability to leverage historical experiences for complex tasks. The framework facilitates dynamic interaction between LLM agents and memory components, supporting multiple LLM backends such as OpenAI's GPT models and Ollama for local deployment. Users can add, read, search, update, and delete memories with ease, while the system automatically evolves memories by finding semantic relationships and updating connections. The project includes detailed usage examples and best practices for memory creation, retrieval, evolution, and error handling. Empirical experiments on six foundation models demonstrate that A-MEM outperforms existing state-of-the-art baselines, highlighting its effectiveness. The project is open-source under the MIT License and provides a citation for academic use. Overall, A-MEM represents a significant advancement in memory systems for LLM agents, enabling more intelligent, adaptive, and context-aware memory management.
https://github.com/agiresearch/A-mem
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.