whereismyai
AutoResearch
AutoResearch is an open-source agent workflow for AI and machine learning research that automates idea generation through paper-ready evidence.
AutoResearch is an open-source agent workflow designed for AI and machine learning research. The system takes research ideas or discovers directions from recent papers and open-source trends, then manages experiment planning, implementation, and analysis. It enables users to iterate based on pilot results and independent reviews. The workflow is stateful and keeps research plans, code, run logs, and critic reports on disk. This structure allows researchers to inspect or take over the process at any stage. It integrates with various LLM providers via API and supports custom local knowledge bases written in Markdown.
AutoResearch at a glance
- Best for: Researcher, Software engineer
- Runs on: Linux, CLI
- Pricing: Free
- Released: August 2026
- Category: AI research tool
- Website: github.com/evomap/autoresearch
Key features
- Cross-domain idea generation from research signals
- Independent multi-model review of generated ideas
- Stateful experiment execution for recoverable workflows
- Pilot testing to check feasibility before scaling
- Traceable recording of experiment results and logs
- Support for documenting negative research results
- Customizable local domain knowledge bases
Use it for
- Discovering research directions from online signals
- Running automated experiments from specific ideas
- Generating paper-ready evidence for AI projects
- Conducting independent multi-model blind reviews
Alternatives to AutoResearch
- OpenResearch — Run parallel research workflows locally or on remote compute.
- Auto Deep Researcher — Automates experiment operations to reduce manual babysitting of training loops.
- HyperResearch — Generates verified reports with citations checked against full source texts and a persistent vault.
- Paper2Agent — It enables direct interaction with research papers and automatic result reproduction.
- AI Scientist v3 — Automates end-to-end scientific research from task descriptions to complete papers.
- AutoRAG — Federates data in place without uploading files to a vector database.
- SoL Pi — Saves $8.75 to $13.50 per hour compared to native Codex and Claude Code systems.
- GPT Researcher — Ranked #1 on Carnegie Mellon's DeepResearchGym benchmark for academic research.
- Local Deep Research — Fully local and encrypted execution with proper citations.
- Arbor — It grows a hypothesis tree and tests changes on held-out data.
- LLM Wiki Research — Compiles parallel agent research into source-backed markdown knowledge bases.
- DeepSearcher — Performs evaluation and reasoning based on private enterprise data using various models.
All AutoResearch alternatives