AI-Scientist vs openevolve
The two are close in size: 14.5k stars for AI-Scientist, 7.3k for openevolve. Over the days we have tracked them AI-Scientist moved +1.6% and openevolve +12.8%, so openevolve is growing faster right now.
openevolve leads on customization and setup ease. AI-Scientist does not take any axis by a clear margin.
Stars and commit dates come from our own daily tracking. The six axes are read off each project's documentation by our review pipeline, so they describe what a project says about itself, not what we measured in its code.
Where they stand today
Automates end-to-end scientific discovery by generating hypotheses, running experiments, and producing full LaTeX papers from templates.
- Stars
- 14.5k
- Tracked growth
- +1.6%
- Maturity
- ●●●●●
- Last commit
- 260d ago
- Language
- Jupyter Notebook
- License
- NOASSERTION
- Cost to run
- Requires your API keys for cloud models; local GPU needed for open-weight runs.
Combines MAP-Elites, island-based evolution and LLM ensembles to autonomously discover novel, hardware-optimized algorithms with reproducible scientific pipelines.
- Stars
- 7.3k
- Tracked growth
- +12.8%
- Maturity
- ●●●●●
- Last commit
- 48d ago
- Language
- Python
- License
- Apache-2.0
- Cost to run
- Your API key; per-iteration LLM costs (local models nearly free)
Six axes, head to head
Each axis runs 0 to 5. The label under a score is what that project's own docs claim, not a category average.
| Axis | AI-Scientist | openevolve |
|---|---|---|
Context depth How much of your codebase it sees before it answers: the open diff, the diff plus related files, or the whole repository. | ●●●●● Template-level view | ●●●●● File + related files |
Noise control How it keeps output volume down — severity thresholds, deduplication, incremental runs over new commits only. | ●●●●● Ensemble reviews | ●●●●● Cascade & novelty filters |
Customization How far it bends to your team: custom rules, prompts, style guides, per-path config. | ●●●●● Template + prompt config | ●●●●● Extensive configs & prompts |
Privacy Whether your code stays on your own infrastructure: fully local, self-hostable, or cloud API only. | ●●●●● Self-hostable (local GPU) | ●●●●● Run fully local |
Model freedom Whether you can point it at any provider, or it is wired to one. | ●●●●● BYOK + local models | ●●●●● Any provider & local |
Setup ease What it takes to get a first useful run out of it. | ●●●●● GPU + dependencies | ●●●●● One-command start |
Which one to pick
Pick AI-Scientist if…
End-to-end experiment automation — pick this when you want a research-focused system that generates ideas, runs experiments on local GPUs, and compiles full papers automatically.
Pick openevolve if…
Research-grade reproducibility — pick this when you need deterministic, scientific-grade autonomous algorithm discovery and hardware-aware code optimization with flexible local or cloud LLM backends.
- Customization: Extensive configs & prompts (5/5 against 3/5)
- Setup ease: One-command start (5/5 against 2/5)
What people want from each one
SakanaAI/AI-Scientist
algorithmicsuperintelligence/openevolve
Questions people ask
Is AI-Scientist better than openevolve?
openevolve leads on customization and setup ease. AI-Scientist does not take any axis by a clear margin. AI-Scientist is worth picking when end-to-end experiment automation — pick this when you want a research-focused system that generates ideas, runs experiments on local GPUs, and compiles full papers automatically.
Which of AI-Scientist and openevolve keeps my code private?
AI-Scientist: Self-hostable (local GPU) (4/5). openevolve: Run fully local (5/5).
What does each one cost to run?
AI-Scientist: Requires your API keys for cloud models; local GPU needed for open-weight runs.. openevolve: Your API key; per-iteration LLM costs (local models nearly free).
Full profiles: SakanaAI/AI-Scientist and algorithmicsuperintelligence/openevolve. Everything else in ML experiment agents.