AGAPI-Agents: An Open-Access Agentic AI Platform for Accelerated Materials Design on AtomGPT.org

2025-12-12

AI Summary

Current AI-driven materials research faces three critical challenges: fragmented computational ecosystems, poor reproducibility, and overreliance on proprietary large language models (LLMs). To address these, we introduce AGAPI—an open-source intelligent agent platform—featuring a novel Agent-Planner-Executor-Summarizer architecture. AGAPI is the first framework to integrate eight or more open-weight LLMs with over twenty domain-specific materials science APIs, unifying access to materials databases, first-principles simulations (tight-binding, machine learning force fields), X-ray diffraction (XRD) analysis, and graph neural networks. It enables end-to-end inverse design workflows exceeding ten reasoning steps and supports multi-tool collaborative inference. Evaluation across 30+ benchmark prompts demonstrates that tool-augmented reasoning significantly improves prediction accuracy. Since launch, AGAPI has attracted over one thousand active users. Its full source code, workflows, and documentation are publicly released, establishing an open, reproducible, and extensible infrastructure for AI-accelerated materials discovery.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Multiagent Systems: Agent/AI Theories and Architectures
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Artificial intelligence is reshaping scientific discovery, yet its use in materials research remains limited by fragmented computational ecosystems, reproducibility challenges, and dependence on commercial large language models (LLMs). Here we introduce AGAPI (AtomGPT.org API), an open-access agentic AI platform that integrates more than eight open-source LLMs with over twenty materials-science API endpoints, unifying databases, simulation tools, and machine-learning models through a common orchestration framework. AGAPI employs an Agent-Planner-Executor-Summarizer architecture that autonomously constructs and executes multi-step workflows spanning materials data retrieval, graph neural network property prediction, machine-learning force-field optimization, tight-binding calculations, diffraction analysis, and inverse design. We demonstrate AGAPI through end-to-end workflows, including heterostructure construction, powder X-ray diffraction analysis, and semiconductor defect engineering requiring up to ten sequential operations. In addition, we evaluate AGAPI using 30+ example prompts as test cases and compare agentic predictions with and without tool access against experimental data. With more than 1,000 active users, AGAPI provides a scalable and transparent foundation for reproducible, AI-accelerated materials discovery. AGAPI-Agents codebase is available at https://github.com/atomgptlab/agapi.

Problem

• Integrates open-source LLMs with materials-science APIs to unify fragmented computational ecosystems
• Autonomously constructs multi-step workflows for materials data retrieval, prediction, and design
• Provides a scalable, transparent platform for reproducible AI-accelerated materials discovery

Innovation

• Integrates open-source LLMs with materials-science APIs
• Uses Agent-Planner-Executor-Summarizer architecture for workflows
• Enables multi-step operations like property prediction and design