EvoScientist

EvoScientist is a self-evolving multi-agent AI system for end-to-end scientific discovery that learns from methods across research runs.

EvoScientist is a living multi-agent AI system designed for end-to-end scientific discovery, helping researchers plan, research, code, analyze, and write. The system features a main agent that composes synchronous and asynchronous workers from specialized agents, general-purpose agents, skills, and MCP tools. It includes a persistent memory that builds a knowledge graph to internalize scholarly taste and support continuous learning across sessions. Users can configure LLM providers, API keys, and models via a setup wizard, and deploy standalone servers via HTTP. Supported inputs include multimodal files such as images, PDFs, and documents with automatic text fallback.

EvoScientist at a glance

Best for: Researcher
Runs on: Web, CLI, API, self-hosted
Pricing: Free
Released: January 2026
Category: AI research tool
Website: evoscientist.ai

Key features

Dynamic multi-agent system
Persistent memory knowledge graph
Deep web search with citation rigor
Sandboxed experiment code execution
Six-phase scientific workflow
Async sub-agents for parallel tasks
Multimodal intake for images and PDFs
Scheduled and autonomous recurring runs

Use it for

Conducting deep web literature research
Writing and debugging experiment code
Running automated scientific workflows
Managing persistent research context

Alternatives to EvoScientist

EvoAgentX — It automatically builds, evaluates, and evolves multi-agent workflows.
DiscoveryRSI — Achieves high actionable-reference rates and supports multi-agent evolutionary research search.
Paper2Agent — It enables direct interaction with research papers and automatic result reproduction.
AI Scientist v3 — Automates end-to-end scientific research from task descriptions to complete papers.
FAROS — Connects literature evidence, executable plans, and human review into an auditable research loop.
OpenResearch — Run parallel research workflows locally or on remote compute.
Arbor — It grows a hypothesis tree and tests changes on held-out data.
Agent Bayes — Builds defensible research mindmaps with page-level citations.
Hypoarena — Provides a fully offline environment to test scientific discovery pipeline logic.
Jevon Tane — It provides a rigorous, pre-registered framework for AI agents to conduct science together.
Biomni — Automates complex wet-lab experimental design with specialized tools and databases.
Auto Deep Researcher — Automates experiment operations to reduce manual babysitting of training loops.

All EvoScientist alternatives