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
zjunlp/KnowAgent
KnowAgent is a project focused on enhancing the planning capabilities of large language model (LLM)-based agents through knowledge augmentation. The core idea is to create an extensive action knowledge base that consolidates action planning knowledge relevant to specific tasks. This knowledge base serves as an external resource that guides the model in generating action sequences or trajectories. The project converts this action knowledge into textual form, enabling the LLM to deeply understand and effectively utilize it for planning complex actions. A key feature of KnowAgent is its knowledgeable self-learning phase, where the model iteratively refines its understanding and application of the action knowledge by learning from the trajectories it generates. This iterative self-improvement process strengthens the agent's planning abilities and adaptability to complex scenarios. The repository provides tools and scripts for planning path generation and self-learning. It supports datasets like HotpotQA and ALFWorld, with scripts to train and test the planning models using different LLMs such as Llama-2. The self-learning process involves converting generated trajectories into a specific format, filtering and merging them based on knowledge, and then training the model further using these refined trajectories. The project also integrates with other open-source tools like FastChat and BOLAA, leveraging their capabilities for training and inference. KnowAgent has been recognized in the research community, with its paper accepted at NAACL 2025 and awarded Best Paper at the KnowledgeNLP workshop at ACL 2024. The project is licensed under Apache 2.0 and aims to provide a robust framework for knowledge-augmented planning in LLM-based agents, facilitating advanced applications in AI planning and decision-making.
https://github.com/zjunlp/KnowAgent
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.