L2 · ANALYST
LLM as Analyst
Research agents with stronger analysis, modeling, and experiment-execution abilities — they plan and reason across steps.
TOOLS IN THIS LEVEL
14 indexedbi
BioAgents
Open-sourceBiology research agents
Research Agents & AI Scientists
bl
BLADE
Open-sourceLLM-assisted data analysis agent
Research Agents & AI Scientists
co
Coscientist
Open-sourceChemistry-focused co-scientist
Research Agents & AI Scientists
di
DiscoveryWorld
Open-sourceScientific discovery benchmark world
Research Agents & AI Scientists
dr
DrugAgent
FreeDrug discovery research agent
Research Agents & AI Scientists
da
DS-Agent
Open-sourceData science research agent
Research Agents & AI Scientists
ds
DSBench
Open-sourceData science agent benchmark
Research Agents & AI Scientists
ls
LLM-SR
Open-sourceLLM for symbolic regression
Research Agents & AI Scientists
ls
LLM-SRBench
Open-sourceBenchmark for LLM symbolic regression
Research Agents & AI Scientists
ml
MLAgentBench
Open-sourceML research agent benchmark
Research Agents & AI Scientists
mb
MLE-bench
Open-sourceML engineer benchmark (OpenAI)
Research Agents & AI Scientists
mc
MLR-Copilot
Open-sourceML research copilot
Research Agents & AI Scientists
ne
NewtonBench
Open-sourcePhysics discovery benchmark
Research Agents & AI Scientists
pr
ProtAgents
Open-sourceProtein research agents
Research Agents & AI Scientists
WHAT DO THE LEVELS MEAN?
- L0 · Reference — Reference Resources: Curated lists, awesome-repos, and initiatives that support AI-in-research workflows — starting points, not end-products.
- L1 · Tool — LLM as Tool: AI tools that augment a single researcher task — literature search, writing, reviewing, figure generation, content transformation.
- L2 · Analyst — LLM as Analyst: Research agents with stronger analysis, modeling, and experiment-execution abilities — they plan and reason across steps.
- L3 · Scientist — LLM as Scientist: End-to-end autonomous research systems. AI-Scientist-style agents that frame hypotheses, run experiments, and write papers.