An LLM agent for end-to-end computational materials discovery

ALKEMIE Agent: an autonomous platform for computational materials design

Despite the powerful multi-scale modeling methods and high-throughput infrastructures established in the materials community, real material computation workflows remain fragmented and heavily manual, requiring researchers to constantly bridge software tools, data analysis, and intermediate decisions. This growing gap between methodological capability and practical execution highlights the need for a new kind of autonomous computational framework, one that can coordinate tools, knowledge, and workflows in a more unified and adaptive way. Here, we introduce ALKEMIE Agent, an agentic platform in which retrieval-augmented generation, a materials-computation knowledge base, registered skills, database-supported provenance, AI-assisted structure modeling, bounded task execution, tool-calling iteration, and error-diagnostic assistance are integrated within a traceable control loop. The capabilities of ALKEMIE Agent are demonstrated through applications including materials recommendation, structure modeling, phonon calculations, machine-learned interatomic potential training, LAMMPS simulations, Ab Initio Monte Carlo (AIMC) sampling, and active-learning-based materials screening. Finally, we outline the future directions and challenges for the development of agentic platforms for computational materials design.

AI agents for MOFs and COFs discovery

Metal-organic frameworks (MOFs) and covalent organic frameworks (COFs) are highly tunable in pore structure and chemical environment, yet their discovery remains slow and fragmented. Synthesis reports are often difficult to compare, characterization data are laborious to interpret, and computational predictions rarely guide experiments directly. Recent advances in large language models (LLM) have enabled the development of artificial intelligence (AI) agents that can interpret research goals, search the literature and databases, call external tools, and adapt workflows based on intermediate results. In this review, we distinguish three stages of AI-agent development in MOFs and COFs research: LLM-native, human-mediated systems; database-grounded, tool-using agents; and experiment-integrated, feedback-driven platforms. This progression reflects increasing scientific grounding and experimental agency. In our view, further progress will depend less on scaling language models alone than on developing traceable machine-actionable data, chemistry-aware validation, persistent experimental memory, and robust interfaces between AI agents and laboratory automation.

Symbolic Predicate-Guided Language Agents for Inverse Design of Perovskite Oxides

This work introduces a domain specific language (DSL)-guided strategy to improve the reasoning and design capability of LLM agents by translating natural language design rules into symbolic predicates encoded in a predefined chemistry DSL, and developed a multi-agent materials design framework.

Human and LLM Collaboration for Accelerated Materials Synthesis and Discovery

Although Large Language Models (LLM) and Artificial Intelligence (AI) tools have enabled a rapid increase in the generation rate of predicted materials, the rate of new materials discovery has lagged behind. This is due to the challenges associated with designing a sequence of chemical reactions to predictably produce new materials, especially in new structure types. Here, we report a study of human and LLM generated recipes for the synthesis of known and new materials. The success of the recipes is determined through in-lab experimentation, and the results are passed back to the humans and LLMs in a closed-loop process to study the effects of their collaboration. The Ruddlesden-Popper homologous series was selected for all material candidates to provide a materials phase space that is simultaneously well studied and likely to host undiscovered materials. We find that humans (H) and LLM (L) have similar success rates: 83(8)% (H) and 75(9)% (L) [known materials, round one], 17(9)% (H) and 22(10)% (L) [unknown materials, round one], 79(8)% (H) and 71(9)% (L) [known materials, round two], and 22(7)% (H) and 14(6)% (L) [unknown materials, round two]. Through this collaborative human-LLM effort, we discovered Ba3PtO5, a material with a new structural prototype that constitutes the missing 1D member of the herein reported dimensionally tunable Rock-Salt Perovskite (RSP) homologous series of the form (AX)m(ABX3)p, of which the Ruddlesden-Popper series is a subset.

Utilizing LLM-based Multi-agent Framework for Pipelining Complex Data Discovery Tasks

From Scientific Copilots to Tool-Grounded Autonomy: AI Agents in Simulation-Driven Materials Discovery

Artificial intelligence (AI) agents and large language model (LLM) agents are beginning to move materials discovery beyond isolated prediction tasks and toward tool-grounded workflows that can retrieve prior knowledge, configure simulations, launch calculations, inspect outputs, and decide what to do next. However, adjacent reviews on materials informatics, self-driving laboratories, natural-language processing in materials science, and autonomous chemistry have not isolated simulation-driven materials workflows as a distinct evidence base. This review addresses that gap through PRISMA-guided searches in Scopus (8 May 2026) and Web of Science (15 June 2026) for English-language journal articles published between 2022 and 2026. The combined search returned 232 records; 27 full texts were assessed and 26 studies were included in the final qualitative synthesis after one full-text exclusion. No eligible study was published in 2022 or 2023, indicating that the field emerged only in 2024 and expanded rapidly in 2025–2026. Catalysis and adsorption tasks (n = 6) and alloy design or evaluation (n = 5) dominated the corpus, while specialized multi-agent architectures were the most common pattern (n = 13). Across the included studies, agentic reasoning was most often coupled to workflow orchestration or integration tools, molecular-dynamics or atomistic simulation environments, and materials-data or machine learning screening pipelines; public repositories or archival artifacts were reported in 18 of 26 studies, experimental validation in six, and robotic closed-loop execution in only one study. The strongest evidence came from workflows that grounded language model decisions in simulators, structured databases, or experimentally verifiable outputs rather than in free-form text alone. This review therefore establishes AI agent workflow orchestration as a distinct analytical category within materials discovery and identifies the reporting, validation, and reproducibility conditions required for these systems to function as credible scientific infrastructure rather than as conversational demonstrations.