AutoResearch vs AI-Scientist-v2
Research agent workflows compared
AutoResearch and AI-Scientist-v2 use AI agents to develop research ideas into experiments and results. AutoResearch emphasizes a traceable, resumable workflow and independent review, while AI-Scientist-v2 explores experiment paths with tree search and can draft papers.
AutoResearch: Turn Research Ideas into Reviewable Evidence
AutoResearch is a Python agent workflow for developing AI and machine learning research ideas into experiments and reviewable evidence. It suits researchers who want a traceable, resumable process and can configure model services and experiment resources.
AI-Scientist-v2: Automate Scientific Experiments and Paper Drafts
AI-Scientist-v2 uses LLM agents and tree search to propose research ideas, run experiments, analyze results, and draft papers. It is aimed at researchers exploring open-ended machine learning questions who can provide model APIs and a controlled compute environment.
| AutoResearch | AI-Scientist-v2 | |
|---|---|---|
| Language | Python | Python |
| License | Apache-2.0 | NOASSERTION |
| Stars | 2.9k | 7.3k |
| Forks | 256 | 995 |
| Last analyzed | Oct 4, 2026 | Oct 3, 2026 |
Key differences
- AutoResearch can start from discovered research directions or a researcher-provided hypothesis; AI-Scientist-v2 generates ideas from a user-provided topic description.
- AutoResearch persists plans, queues, logs, results, and decisions, and includes pilot experiments and independent review; AI-Scientist-v2 uses best-first tree search to explore experiment alternatives and produce manuscript drafts.
- AutoResearch's documented workflow requires Bun, Node.js, an experiment environment manager, and experiment-specific resources; AI-Scientist-v2 setup targets Linux with NVIDIA GPUs, CUDA, and PyTorch.
- AutoResearch lists an Apache-2.0 license; AI-Scientist-v2 lists NOASSERTION in its metadata and describes a project-specific source code license with a manuscript disclosure requirement.
- AutoResearch reports 2.9k stars and 256 forks, while AI-Scientist-v2 reports 7.3k stars and 995 forks.
Choose AutoResearch if you…
- want durable experiment records that can be resumed and examined.
- need pilot experiments, preserved negative results, and independent review in the workflow.
- can configure multiple model roles and the required experiment environment.
Choose AI-Scientist-v2 if you…
- want to explore experiment alternatives through progressive tree search.
- need a pipeline that can produce an initial manuscript draft from experiment results.
- can provide supported model APIs and run LLM-generated code in a controlled compute environment.
Compare the numbers side by side, or add more projects
This comparison is generated with AI from the OSRepos analyses of both projects. Always check each project's repository and documentation before choosing.