Skip to main content
Back to trending

AI-Scientist-v3

The Agentic Scientist: Orchestrating Coding Agents for Automated Scientific (Co-)Discovery

View on GitHub
Stars
304
Forks
49
+ today
+2
Created
16d

Ranking data as of October 26, 2026 (UTC).

Star History

Today, hour by hour

304 stars
302304

Overview

The Agentic Scientist: Orchestrating Coding Agents for Automated Scientific (Co-)Discovery It ranks #321 on GitTiger, gaining +2 stars on October 26, 2026 (UTC).

The project is written in Python and has 49 forks. It was created 16d ago.

AI-generated summary of the repository README — may contain errors.

What it is

AI-Scientist-v3 is a Python project under the Apache-2.0 license, per the supplied metadata. Its description calls it an agentic scientist that orchestrates coding agents for automated scientific (co-)discovery. The supplied topics list is empty.

Capabilities

The README preview adds these details: - A pipeline that runs coding agents for scientific work. The preview is truncated, so the full pipeline is not supplied. - Agent CLIs packaged inside a container image, with a native (non-containerised) route also described. - An optional Autoillustrate component, built into the image, for concept figures. - GPUs are needed only by the GPU task presets.

Getting started

The preview lists these prerequisites: Python 3.11 or newer, with uv. It says API keys belong in a .env file at the repository root. No commands were supplied, so follow the README.

Installation
git clone https://github.com/SakanaAI/AI-Scientist-v3.git
cd AI-Scientist-v3
# see README for setup

Frequently Asked Questions

Which Python version is required?

The README preview lists Python 3.11 or newer, with uv.

Do I need a GPU?

The preview says GPUs are required only by the GPU task presets.

Where do API keys go?

The preview says they go in a .env file at the repository root.