EveryDev.ai
Subscribe
Home
Tools

5,923+ AI tools

  • Featured
  • Gems
  • Trending
  • New
  • Rate tools
  • Compare
  • Arena
Categories
  • Agents2992
  • Coding2260
  • Infrastructure1068
  • Projects824
  • MCP795
  • Research744
  • Analytics567
  • Marketing527
  • Design489
  • Testing476
  • Data442
  • Security417
  • Integration288
  • Communication286
  • Prompts284
  • Extensions239
  • Voice214
  • DevOps191
  • Commerce184
  • Web128
  • Learning123
  • Finance44
AI Tools by Topic
  • AI Coding Assistants
  • Agent Frameworks
  • MCP Servers
  • AI Prompt Tools
  • Vibe Coding Tools
  • AI Design Tools
  • AI Database Tools
  • AI Website Builders
  • AI Testing Tools
  • LLM Evaluations
Follow Us
  • X / Twitter
  • LinkedIn
  • Reddit
  • Discord
  • Threads
  • Bluesky
  • Mastodon
  • YouTube
  • GitHub
  • Instagram
Get Started
  • Users
  • Rate Tools
  • About
  • Editorial Standards
  • Corrections & Disclosures
  • Community Guidelines
  • Advertise
  • Contact Us
  • Newsletter
  • Submit a Tool
  • Start a Discussion
  • Write A Blog
  • Share A Build
  • Terms of Service
  • Privacy Policy
Explore with AI
  • ChatGPT
  • Gemini
  • Claude
  • Grok
  • Perplexity
Agent Experience
  • llms.txt
Theme
With AI, Everyone is a Dev. EveryDev.ai ©
    1. Home
    2. Tools
    3. The Agentic Scientist (AI-Scientist-v3)
    The Agentic Scientist (AI-Scientist-v3) icon

    The Agentic Scientist (AI-Scientist-v3)

    Autonomous Systems
    Featured

    Open-source pipeline that orchestrates coding agents to turn a task description into experiments, ablations, and a written paper.

    Visit Website

    At a Glance

    Pricing
    Open Source

    Self-hosted core under the Apache License 2.0. Users supply their own API keys and compute (external costs).

    Engagement

    Available On

    Windows
    Linux
    API
    CLI

    Resources

    WebsiteDocsGitHubllms.txt

    Topics

    Autonomous SystemsMulti-agent SystemsResearch Assistants

    Alternatives

    NanoResearchScientific Agent SkillsMiroThinker
    Developer
    Sakana AIAzabudai Hills Mori JP Tower 22F, Minato-kuEst. 2023$347M+ raised

    Listed Nov 2026

    About The Agentic Scientist (AI-Scientist-v3)

    The Agentic Scientist, from Sakana AI, is an open-source pipeline that treats coding agents as AI scientists. Starting from a task.md, agents propose ideas, run experiments, evaluate results against a task rubric, and iterate, then the pipeline runs baselines, ablations, a sealed final evaluation, post-hoc analyses, and writes the paper.

    What It Is

    Agentic Scientist is a command-line orchestration framework (agsci) for automated scientific discovery. It drives coding agent CLIs such as claude, codex, cursor, opencode, gemini, and pi through Headless CLI, so the same experiment can be repeated across different agents and models. Every agent phase runs inside Docker by default.

    How the Pipeline Works

    Six phases run in sequence: task drafting, investigation, ablations, final evaluation, post-hoc analysis, and paper writing, with optional review and revision phases afterwards. Investigation is the idea, experiment, evaluate loop. Ablations compare the winning program against baselines and ablated variants, and the final evaluation re-scores it in a sealed environment. Each phase reads only a run-local snapshot of the task, so the shipped examples/ are never modified.

    Outputs and Inspection

    Each run writes a single directory under results/<task>/<agent>/<run_id>/ containing drafting logs, the task snapshot, investigation records, ablations, final evaluation, post-hoc analyses, and the paper (manuscript, PDF, manifest, and optional AI reviews). Runs can be browsed in a local web UI with agsci inspect start. An OpenAI-backed reviewer can also be run on any PDF with agsci review.

    Setup Path and Safety

    Setup involves building the Docker agent image, installing the Python environment with uv, and supplying agent API keys in a .env file. agsci doctor checks the host and --dry-run previews commands. The CPU-only circle-packing example is the recommended first run; a GRPO example requires GPUs. The README warns that agents run with full tool permissions and may inherit API keys, so runs should stay inside Docker and only trusted tasks should be used. Drafting a task from a bare research brief is described as experimental.

    The Agentic Scientist (AI-Scientist-v3) - 1

    Community Discussions

    Be the first to start a conversation about The Agentic Scientist (AI-Scientist-v3)

    Share your experience with The Agentic Scientist (AI-Scientist-v3), ask questions, or help others learn from your insights.

    Pricing

    OPEN SOURCE

    Open Source (Apache 2.0)

    Self-hosted core under the Apache License 2.0. Users supply their own API keys and compute (external costs).

    • Apache License 2.0
    • Runs agent phases in Docker via agsci CLI
    • Supports claude, codex, cursor, opencode, gemini, and pi agents via Headless CLI
    • Requires your own API keys (BYOK) such as ANTHROPIC_API_KEY, OPENAI_API_KEY, GEMINI_API_KEY
    • GPUs required only by GPU task presets

    Capabilities

    Key Features

    • Six-phase pipeline: task drafting, investigation, ablations, final evaluation, post-hoc analysis, paper
    • Works with claude, codex, cursor, opencode, gemini, and pi agents via Headless CLI
    • Docker-contained agent execution
    • Sealed final evaluation and reproducible run-local task snapshots
    • Baselines and ablation studies
    • Automated paper writing and compilation with optional AI review
    • Resume runs or run selected phases
    • Run Inspector web UI
    • Hydra key=value configuration and task presets
    • GPU task presets
    • Optional Autoillustrate concept figures

    Integrations

    Claude
    OpenAI Codex
    Cursor
    OpenCode
    Gemini
    Pi
    Headless CLI
    Docker
    Autoillustrate
    Amazon Bedrock
    API Available
    View Docs

    Ratings & Reviews

    No ratings yet

    Be the first to rate The Agentic Scientist (AI-Scientist-v3) and help others make informed decisions.

    Rate other tools you’ve used

    Developer

    Sakana AI

    Sakana AI builds nature-inspired AI research systems, combining evolutionary algorithms and large language models to automate scientific discovery. The team, led by researchers with backgrounds at Google DeepMind and other leading AI labs, develops open-source frameworks like ShinkaEvolve, the AI Scientist, and the Darwin Goedel Machine. Sakana AI publishes peer-reviewed research and releases production-ready tools for the scientific computing community.

    Founded 2023
    Azabudai Hills Mori JP Tower 22F, Japan
    $347M+ raised
    218 employees

    Used by

    MUFG Bank
    SMBC Group
    Daiwa Securities Group
    Ministry of Defense of Japan
    +3 more
    Read more about Sakana AI
    WebsiteGitHubX / Twitter
    4 tools in directory

    Similar Tools

    NanoResearch icon

    NanoResearch

    Open-source end-to-end autonomous AI research engine that turns a research topic into experiments, figures and a LaTeX paper.

    Scientific Agent Skills icon

    Scientific Agent Skills

    An open-source library of 168 ready-to-use scientific and research skills for AI agents, covering biology, chemistry, medicine, drug discovery, and 100+ scientific databases.

    MiroThinker icon

    MiroThinker

    MiroThinker is an open-source deep research agent series by MiroMind, optimized for complex multi-step reasoning, web research, and prediction tasks with state-of-the-art benchmark performance.

    Browse all tools

    Related Topics

    Autonomous Systems

    AI agents that can perform complex tasks with minimal human guidance.

    561 tools

    Multi-agent Systems

    Platforms for creating and managing teams of AI agents that can collaborate.

    621 tools

    Research Assistants

    AI assistants specialized in helping with research tasks.

    110 tools
    Browse all topics
    Back to all toolsSuggest an edit
    ratings
    discussions
    15views