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
evo-hq/evo
evo is a Python CLI and agent plugin that turns a codebase into an automated research loop. Given a repository, the discover skill explores it, asks or infers what should be optimised, which command runs as the benchmark and which direction the metric should move, and instruments the evaluation, adding a held-out-slice score floor gate when it builds a benchmark from scratch. The optimize command then runs the loop: experiments are proposed, applied and scored, improvements are kept and regressions discarded. The design extends the greedy hill climb of Karpathy autoresearch with several mechanisms. Search is a tree rather than a single branch, so multiple directions can fork from any committed node. Subagents run in parallel, each in its own git worktree, reading shared state that holds failure traces, annotations and discarded hypotheses before forming a hypothesis and running the benchmark, and each may take several iterations within its branch. After every round a frontier strategy chooses which committed branch to extend next, with argmax, top_k, epsilon_greedy, softmax and a Pareto per-task option that preserves specialists an aggregate score would hide. Between rounds, scan subagents read batches of traces in parallel and surface compound failure patterns such as gate-failure intersections and shared root causes. Gates are pass or fail checks, any command exiting zero on success, that discard an experiment even when its score improves, which prevents the search from returning constants, skipping work or trading correctness for speed; gates inherit down the tree and narrower gates can attach to specific branches. evo installs as a uv tool with per-host plugins for Claude Code, Codex, Cursor, Kimi, OpenClaw, Hermes, Opencode and Pi, runs experiments in local worktrees, a workspace pool, over SSH or on Modal, E2B, Daytona, AWS and Azure, and ships a dashboard for monitoring.
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.