automated scientific discovery / AI research agents
AI-Scientist-v2 is an autonomous agentic system that performs end-to-end scientific discovery via agentic tree search, automatically generating hypotheses, running experiments, analyzing data, and writing papers.
Solves the same problem with a different architecture or at a different layer. Expect to rewrite the integration.
choose it if you want self-organizing agent teams for long-running experiments rather than Sakana's tree-search stack.
AutoScientists: Self-Organizing Agent Teams for Long-Running Scientific Experimentation
choose it if you want to run an implementation of the Kosmos autonomous-discovery paper (via Claude Code/API) instead of v2.
Kosmos: An AI Scientist for Autonomous Discovery - An implementation and adaptation to be driven by Claude Code or API - Based on the Kosmos AI Paper - https://arxiv.org/abs/2511.02824
choose it if you want a chat-initiated, self-evolving research loop instead of v2's tree-search pipeline.
Fully autonomous & self-evolving research from idea to paper. Chat an Idea. Get a Paper. 🦞
choose it if you want a human-in-the-loop multi-agent research workflow from implementation to paper.
Agent Laboratory is an end-to-end autonomous research workflow meant to assist you as the human researcher toward implementing your research ideas
choose it if you prefer blueprint-driven orchestration of AI research workflows over agentic tree search.
A blueprint-driven AutoResearch runtime for orchestrating AI research workflows from idea generation and experiments to paper writing and peer review.
choose it if you want a compact, assistant-oriented research agent instead of the tree-search harness.
🦞+🔬 NanoResearch: The Autonomous AI Research Assistant
choose it if you want a unified long-horizon agentic framework for scientific discovery from a different research group.
InternAgent-1.5: A Unified Agentic Framework for Long-Horizon Autonomous Scientific Discovery
choose it if you want the AI-Researcher framework (NeurIPS 2025) for autonomous scientific innovation rather than v2's tree search.
[NeurIPS2025] "AI-Researcher: Autonomous Scientific Innovation" -- A production-ready version: https://novix.science/chat
An earlier or more limited way to do the job. Still the right call when you need something small, proven, or CPU-only.
choose v1 when you have a clear template and want higher paper-success rates for well-defined tasks.
The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery 🧑🔬
choose it if you need an autonomous research loop that runs natively on Apple Silicon (MLX) without PyTorch.
Apple Silicon (MLX) port of Karpathy's autoresearch — autonomous AI research loops on Mac, no PyTorch required.
choose it if you need a narrower, data-to-manuscript pipeline with backward traceability from paper to code/data.
data-to-paper: Backward-traceable AI-driven scientific research
Does the same job, but ships as an app. Useful to a person, not swappable into a codebase.
choose it if you want an interactive AI workbench (web/desktop) for managing research, rather than a code-level pipeline.
The open-source AI workbench for scientific research
These projects were analysed and named AI-Scientist-v2 among their alternatives. The relationship is not symmetric — how AI-Scientist-v2 rates them is a separate judgement, made when AI-Scientist-v2 is analysed in its own right.
calls AI-Scientist-v2 “Direct successor with agentic tree search; choose for the improved workshop-level version.”
The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery 🧑🔬
calls AI-Scientist-v2 “choose for agentic tree search over a full autonomous discovery system rather than skills”
Academic Research Skills for Claude Code: research → write → review → revise → finalize
calls AI-Scientist-v2 “Choose for full automated scientific discovery via tree search instead of composing individual skills.”
Turn any AI agent into an AI Scientist. The #1 Agent Skills library for science, used by 190,000+ scientists worldwide. 165 ready-to-use validated skills plus 100+ scientific databases covering biology, chemistry, medicine, and drug discovery. Compatible with Cursor, Claude Code, Codex, Pi, Antigravity, and the open Agent Skills standard.
calls AI-Scientist-v2 “choose it if you want the improved v2 with agentic tree search for deeper exploration.”
Comprehensive open-source library of AI research and engineering skills for any AI model. Package the skills and your claude code/codex/gemini agent will be an AI research agent with full horsepower. Maintained by Orchestra Research.
calls AI-Scientist-v2 “Agentic tree-search variant of the same autonomous pipeline; a heavyweight Python system you execute rather than skills you install.”
An AI workflow skill pack for research, competitions, and innovation projects.