Autonomous Science Agents — Researcher Index

v2.3.0 · updated 2026-10-10
A filterable index of AI systems that act as scientific agents: literature, hypothesis, experiment planning, code/simulation execution, lab automation, analysis, and reporting across STEM. Answer fast: what domain? · in → out? · accessible today? · how autonomous? Local 2.3.0 preview reviewed 2026-10-10. 542 records; 84 carry this review date. 84 records were added or restored after an independent adversarial check. 76 existing records received targeted corrections, mostly access relabels after a licence check, and keep their earlier verification dates. The October audit records sources, review verdicts, holds and coverage limits. Access tags are best-effort — verify before relying on any system. Rows tagged Paper-only are leads from papers with no confirmed public code. Click a row for detail.
Dedicated to ELIXIR-CZ
Czech national node of ELIXIR
542systems & benchmarks
262open source
55physical-lab (A5)
134benchmarks
5domains
Domain
Access
Autonomy
Only
⚡ Best starting points — by need
Biomedical agent you can runBiomni, PaperQA2, TxAgent
Literature-grounded QAPaperQA2, Asta, Elicit
End-to-end ML researchThe AI Scientist, AI Scientist-v2, Agent Laboratory
Autonomous discovery platformKosmos, Google AI Co-Scientist, Robin
Single-cell / omics analysisCellVoyager, SpatialAgent, CellAgent
Chemistry tool-use agentChemCrow, CACTUS, ChemToolAgent
Self-driving chemistry labCoscientist, Chemist-X, A-Lab
Materials / DFT automationAtomAgents, VASPilot, MatClaw, ChemGraph
Physics / simulationOpenFOAMGPT, MechAgents, cmbagent
Protein / antibody designProtAgents, The Virtual Lab, OriGene
Evaluate an agent (benchmark)ScienceAgentBench, AstaBench, MLE-bench
Inputs → OutputsLinks