FAROS

FAROS is a collaborative multi-agent system that orchestrates AI research workflows from ideas and experiments to papers and reviews.

FAROS is a multi-agent system built around the real scientific workflow. It connects research ideation, literature retrieval, planning, code generation, experiment tracking, paper writing, and peer review while preserving provenance, decisions, artifacts, and human feedback at every stage. The current release focuses on LLM and AI Scientist research. Its runtime follows a Blueprint, Capability, Profile, and Provider design so that new workflows, tools, and scientific domains can be added. It uses PlanPackage as a typed handoff contract to carry scientific intent across modules, and ReviewX to check claim, evidence, and measurement consistency.

FAROS at a glance

Best for: Researcher
Runs on: Web, Python, Docker, self-hosted, open source
Pricing: Free
Released: May 2026
Category: AI research tool
Website: github.com/OpenNSWM-Lab/FAROS

Key features

Run literature retrieval and semantic filtering before ideation
Use PlanPackage as a typed cross-module handoff contract
Generate code, run sandbox execution, and evaluate results
Produce publication-grade figures and LaTeX or PDF papers
Run ReviewX for claim-evidence-measurement consistency checks
Support human-governed automation and approvals
Recover long-running background tasks after browser disconnects
Provide user-scoped providers and encrypted API keys

Use it for

Refine rough research interests into executable topics
Generate evidence-supported research plans and hypotheses
Execute code and collect metrics in sandboxed environments
Draft papers and generate LaTeX or PDF documents
Run peer review and generate actionable revision plans

Alternatives to FAROS

AI Scientist v3 — Automates end-to-end scientific research from task descriptions to complete papers.
Fara — It operates visually to perform multi-step web tasks without accessibility trees.
OpenResearch — Run parallel research workflows locally or on remote compute.
Paper2Agent — It enables direct interaction with research papers and automatic result reproduction.
Lune — Accesses trusted knowledge from top conference papers for AI agents.
Agent Bayes — Builds defensible research mindmaps with page-level citations.
Rayyan — Automates screening and data extraction to accelerate systematic reviews.
BrowserOS — Runs entirely locally and lets AI agents use real user logins to complete tasks.
MiroFlow — Achieves top rankings on multiple agentic benchmarks with modular architecture.
Hypoarena — Provides a fully offline environment to test scientific discovery pipeline logic.
DiscoveryRSI — Achieves high actionable-reference rates and supports multi-agent evolutionary research search.
Jevon Tane — It provides a rigorous, pre-registered framework for AI agents to conduct science together.

All FAROS alternatives