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
Optima-CityU/LLM4AD_Next
LLM4AD_Next is a platform for automated algorithm design driven by large language models, developed by the Optima group at City University of Hong Kong as the successor to the earlier LLM4AD project. Its purpose is to remove the manual configuration pipeline that such systems normally require. An interactive conversational terminal interviews the user about the optimisation problem at hand and then generates the whole runnable scaffold: an evaluator, an algorithm skeleton, a configuration file and a debugger. From there the system runs an evolutionary loop in which an LLM proposes and mutates candidate heuristics in code, the evaluator scores them on the problem instances, and the population improves over generations. Several automatic heuristic design search methods are available as standalone orchestrators, including EoH, MEoH, ReEvo, MCTS-AHD, IslandGA and DyCA, selected by a single field in the YAML configuration; further methods such as FunSearch, LLaMEA and NSGA-II are listed as pending migration from the original platform. An evolve-block advisor can be pointed at an existing repository, where it scans the code, scores blocks and recommends which ones are most promising to evolve toward a stated goal. Long-term memory is provided through an external memory service with global, project and task scopes and configurable chat and embedding model bindings. The project ships versioned Docker images aligned with GitHub release tags for reproducible deployment, offers a browser-based trial that needs no local setup or API key, and installs from PyPI for Python 3.12 and newer under a BSD-3-Clause licence. It targets researchers and practitioners in optimisation, operations research and heuristic design who want to apply LLM-driven program evolution, in the style of FunSearch and AlphaEvolve, without assembling the pipeline themselves.
https://github.com/Optima-CityU/LLM4AD_Next
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.