Research teams

Working harness designs — topology, agent roster, model choices, and the evidence behind them. 18 teams match.

Task kind

SortFeatured Recently active Highest trust

Agent Laboratory

Samuel Schmidgall

Self-ReportedClaims stated by the owner. 2 of the 3 evidence-linked proof entries needed for Evidence-Linked are on record.Curated

Research assistants for literature review, experiment execution and report preparation.

updated 18d ago

Pipeline5 listed membersCustom

Literature researcherExperiment plannerExperiment implementerData preparation contributor+1 more

research

Outcome[unknown]

Economics[unknown] · deliberate

2 proof

🕸

AI-Researcher

HKUDS

Self-ReportedClaims stated by the owner. 2 of the 3 evidence-linked proof entries needed for Evidence-Linked are on record.Curated

A research pipeline evaluated on scientific implementation and manuscript tasks.

updated 18d ago

Pipeline5 listed membersCustom

Research resource collectorResearch analystResearch implementerImplementation adviser+1 more

research

Outcome[unknown]

Economics[unknown] · deliberate

2 proof

🕸

The AI Scientist-v2

Sakana AI

Self-ReportedClaims stated by the owner. 2 of the 3 evidence-linked proof entries needed for Evidence-Linked are on record.Curated

An autonomous research system generated experimental work and manuscripts for a workshop review study.

updated 18d ago

Other1 listed memberCustom

Experiment search guidance

research

Outcome[unknown]

Economics[unknown] · deliberate

2 proof

🕸

Virtual Lab

zou-group

Self-ReportedClaims stated by the owner. 2 of the 3 evidence-linked proof entries needed for Evidence-Linked are on record.Curated

A human-guided research team used to design SARS-CoV-2 nanobodies.

updated 18d ago

Supervisor4 listed membersCustom

Research leadImmunology adviserMachine learning specialistComputational biology specialist

research

Outcome[unknown]

Economics[unknown] · deliberate

2 proof

🌐

AgentVerse Dynamic Group

Tsinghua / BUPT / Tencent WeChat AI — OpenBMB (AgentVerse)

Self-ReportedClaims stated by the owner. 1 of the 3 evidence-linked proof entries needed for Evidence-Linked is on record.Curated

Dynamically recruited specialist group — outperforms single agents on reasoning, coding and tool-use tasks.

updated 3y ago

Supervisor1 listed memberAgentVerse

Recruited Specialist

researcheducation

Outcome89%

Economics[unknown] · deliberate

1 proof

🕸

AI co-scientist

Google Research

Self-ReportedClaims stated by the owner. 1 of the 3 evidence-linked proof entries needed for Evidence-Linked is on record.Curated

Specialized agents develop and refine hypotheses for scientific investigations.

updated 18d ago

Supervisor7 listed membersCustom

CoordinationScientific reasoningScientific reasoningScientific reasoning+3 more

research

Outcome[unknown]

Economics[unknown] · deliberate

1 proof

🔀

Anthropic Orchestrator-Workers Pattern

Anthropic

Self-ReportedClaims stated by the owner. 1 of the 3 evidence-linked proof entries needed for Evidence-Linked is on record.Curated

Central LLM dynamically breaks down tasks and delegates to specialist workers.

updated 1y ago

Orchestrator–Worker2 listed membersClaude API

OrchestratorWorker

software-deliveryresearchdata-extraction

Outcome[unknown]

Economics[unknown] · deliberate

1 proof

💬

AutoGen Group Chat

Microsoft Research (AutoGen)

Self-ReportedClaims stated by the owner. 1 of the 3 evidence-linked proof entries needed for Evidence-Linked is on record.Curated

Flexible multi-agent group conversation — hierarchical, peer, or proxy topologies.

updated 3y ago

Supervisor1 listed memberAutoGen

ConversableAgent

software-deliveryresearchdata-extraction

Outcome69.5%

Economics[unknown] · deliberate

1 proof

🐪

CAMEL Role-Playing Two-Agent

CAMEL-AI (King Abdullah University of Science and Technology)

Self-ReportedClaims stated by the owner. 1 of the 3 evidence-linked proof entries needed for Evidence-Linked is on record.Curated

Inception-prompted AI User + AI Assistant — autonomous cooperative task completion.

Economics[unknown] · deliberate

1 proof

🕸

Claude Research

Anthropic

Self-ReportedClaims stated by the owner. 1 of the 3 evidence-linked proof entries needed for Evidence-Linked is on record.Curated

Parallel research agents gather information for a lead researcher and citation agent.

updated 1y ago

Orchestrator–Worker3 listed membersCustom

Research coordinationResearchCitation placement

research

Outcome[unknown]

Economics[unknown] · deliberate

1 proof

🚢

CrewAI Sequential Research Crew

CrewAI

Self-ReportedClaims stated by the owner. 1 of the 3 evidence-linked proof entries needed for Evidence-Linked is on record.Curated

Researcher → Analyst sequential pipeline with natural context handoff.

updated 2y ago

Pipeline2 agentsCrewAI

ResearcherAnalyst

researchcontent

Outcome[unknown]

Economics[unknown] · deliberate

1 proof

🏘️

Generative Agents Sandbox

Park et al. — Stanford University, with Google Research and Google DeepMind

Self-ReportedClaims stated by the owner. 1 of the 3 evidence-linked proof entries needed for Evidence-Linked is on record.Curated

25 autonomous NPC agents with memory, reflection, planning — emergent social org.

updated 3y ago

Swarm25 agentsCustom (The Sims-inspired sandbox)

Smallville Resident

researchgaming

Outcome[unknown]

Economics[unknown] · deliberate

1 proof

🕹️

Magentic-One

Microsoft Research

Self-ReportedClaims stated by the owner. 1 of the 3 evidence-linked proof entries needed for Evidence-Linked is on record.Curated

Microsoft Research generalist 5-agent system: GAIA 32.33%, WebArena 32.8%.

updated 1y ago

Supervisor5 agentsAutoGen

Orchestrator· GPT-4oWebSurfer· GPT-4oFileSurfer· GPT-4oCoder· GPT-4o+1 more

researchsoftware-deliverydata-extraction

Outcome[unknown]

Economics[unknown] · deliberate

1 proof

📊

MALBO Bayesian-Optimized Team

Antonio Sabbatella (MALBO master's thesis, University of Milano-Bicocca)

Self-ReportedClaims stated by the owner. 1 of the 3 evidence-linked proof entries needed for Evidence-Linked is on record.Curated

Multi-objective Bayesian search for team config — >45% cost reduction vs random.

updated 11mo ago

Supervisor5 agentsMALBO

Optimized Team Member

researchsoftware-delivery

Outcome[unknown]

Economics[unknown] · deliberate

1 proof

🕸

Robin

FutureHouse

Self-ReportedClaims stated by the owner. 1 of the 3 evidence-linked proof entries needed for Evidence-Linked is on record.Curated

A scientific workflow connects literature research and data analysis to human-run experiments.

updated 18d ago

Other3 listed membersCustom

Literature researchCandidate evaluationData analysis

research

Outcome[unknown]

Economics[unknown] · deliberate

1 proof

🐍

smolagents CodeAgent

Hugging Face

Self-ReportedClaims stated by the owner. 1 of the 3 evidence-linked proof entries needed for Evidence-Linked is on record.Curated

Python-first multi-provider agent — minimal 1K-line library, MCP + LangChain tools.

updated 1y ago

Solo + Tools1 agentsmolagents (Python)

CodeAgent

software-deliveryresearchdata-extraction

Outcome[unknown]

Economics[unknown] · deliberate

1 proof

⛏️

Voyager (Minecraft)

NVIDIA Research (Voyager)

Self-ReportedClaims stated by the owner. 1 of the 3 evidence-linked proof entries needed for Evidence-Linked is on record.Curated

Open-ended Minecraft agent — 3.3× items; tech tree up to 15.3× faster (wooden tool level) vs prior SOTA.

updated 3y ago

Solo + Tools1 agentGPT-4 API (Mineflayer/Minecraft)

Explorer· GPT-4

gamingresearch

Outcome[unknown]

Economics[unknown] · deliberate

1 proof

🌅

Helios Swarm

Helios Labs

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Illustrative

Homogeneous extraction worker pool deployed from a single blueprint.

updated 5mo ago

Swarm3 agentsCustom

Blueprint origin· Claude Haiku 4.5Worker· Claude Haiku 4.5Worker· Claude Haiku 4.5

data-extractionresearch

Outcome[unknown]

Economics2.1M

0 proof