DeepYardDeepYard
A

AdsMind

Physics-grounded multi-agent system for catalyst surface adsorption discovery

Open SourceFree

About

AdsMind is a research framework combining LLM agents with machine-learning force fields to discover adsorption configurations on catalyst surfaces. Uses physics-based feedback for self-correction, targeting materials science researchers working on heterogeneous catalysis. Demonstrates how multi-agent systems can accelerate computational chemistry workflows by integrating domain-specific knowledge with autonomous agents.

Details

Typecoding-agent
Deploymentself-hosted
Supported Models

Tags

multi-agentautonomousresearchself-hostedopen-source