autonomous experiment loops / autoresearch
Autonomous experiment loop extension for the pi terminal AI coding agent that enables try-measure-keep/revert optimization loops.
Solves the same problem with a different architecture or at a different layer. Expect to rewrite the integration.
standalone idea-to-paper research agent, choose if you want a self-contained pipeline rather than a pi extension.
Fully autonomous & self-evolving research from idea to paper. Chat an Idea. Get a Paper. 🦞
end-to-end autonomous research workflow, choose if you want a full research assistant outside pi.
Agent Laboratory is an end-to-end autonomous research workflow meant to assist you as the human researcher toward implementing your research ideas
blueprint-driven AutoResearch runtime, choose if you want a standalone orchestrator for idea-to-paper.
A blueprint-driven AutoResearch runtime for orchestrating AI research workflows from idea generation and experiments to paper writing and peer review.
choose if you need the original autoresearch loop on Apple Silicon without pi or PyTorch.
Apple Silicon (MLX) port of Karpathy's autoresearch — autonomous AI research loops on Mac, no PyTorch required.
standalone tool that auto-instruments codebase benchmarks, choose if you don't use pi.
turns your codebase into an autoresearch loop — discovers what to measure, instruments the benchmark, then runs tree search with parallel subagents.
An earlier or more limited way to do the job. Still the right call when you need something small, proven, or CPU-only.
narrower: focused only on deep-learning experiment runs, not general optimization targets.
🔥 An autonomous AI agent that runs your deep learning experiments 24/7 while you sleep. Zero-cost monitoring, Leader-Worker architecture, constant-size memory.
narrower: Claude Code skill optimizer rather than a general pi experiment loop.
达尔文.skill —— 一个让你的Skill无限进化的系统：评估→改进→测试→保留或回滚 | Autoresearch-inspired autonomous skill optimization for Claude Code. Evaluate, improve, test, keep or revert.
narrower: Claude Code skills for ML research, not a pi-general loop.
ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.
These projects were analysed and named pi-autoresearch among their alternatives. The relationship is not symmetric — how pi-autoresearch rates them is a separate judgement, made when pi-autoresearch is analysed in its own right.
calls pi-autoresearch “Autonomous experiment loop built specifically for the pi agent; use it if pi is your host.”
ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.
calls pi-autoresearch “choose it if you use the pi agent and need a specialized autonomous experiment loop.”
Comprehensive open-source library of AI research and engineering skills for any AI model. Package the skills and your claude code/codex/gemini agent will be an AI research agent with full horsepower. Maintained by Orchestra Research.
calls pi-autoresearch “An experiment-loop extension living inside the pi agent; pick it for narrow experiment automation inside pi only.”
The most autonomous AI agent framework: runs unattended on GitHub Actions, self-healing skills, drives Claude Code, Grok, Codex & more. No approval loops. Configure once, forget forever.