El Agente Cuántico: Automating quantum simulations

TL;DR

El Agente Cuántico introduces a multi-agent AI system that translates natural-language prompts into end-to-end quantum-simulation workflows across a heterogeneous software stack. By grounding reasoning in library documentation and APIs, it autonomously assembles and validates tasks spanning state preparation, static and dynamic simulations, tensor-network methods, quantum control, error correction, and resource estimation. The paper demonstrates a broad set of experiments—ranging from VQE and Bell-state benchmarks to HEOM, Floquet dynamics, TEBD/TD-DMRG, phase diagrams, and QEC—to illustrate end-to-end autonomy and cross-platform integration. It also presents a pragmatic roadmap toward progressively higher autonomy, including multi-backend orchestration, hybrid classical-quantum workflows, hardware-awareness, and eventual self-driving discovery. The work highlights both the potential to accelerate quantum-science workflows and the dependency on accurate, up-to-date documentation and robust tool ecosystems.

Abstract

Quantum simulation is central to understanding and designing quantum systems across physics and chemistry. Yet it has barriers to access from both computational complexity and computational perspectives, due to the exponential growth of Hilbert space and the complexity of modern software tools. Here we introduce{\cinzel El Agente Cuántico}, a multi-agent AI system that automates quantum-simulation workflows by translating natural-language scientific intent into executed and validated computations across heterogeneous quantum-software frameworks. By reasoning directly over library documentation and APIs, our agentic system dynamically assembles end-to-end simulations spanning state preparation, closed- and open-system dynamics, tensor-network methods, quantum control, quantum error correction, and quantum resource estimation. The developed system unifies traditionally distinct simulation paradigms behind a single natural-language interface. Beyond reducing technical barriers, this approach opens a path toward scalable, adaptive, and increasingly autonomous quantum simulation, enabling faster exploration of physical models, rapid hypothesis testing, and closer integration between theory, simulation, and emerging quantum hardware.

Figures (16)

• Figure 1: Schematic overview of El Agente Cu√°ntico, detailing its capabilities (upper half) and the integrated high-performance software stack and tools (bottom half).
• Figure 2: Multiagentic architecture of El Agente Cuántico. The central agent orchestrates different experts capable of designing and executing code using a specific quantum simulation package.
• Figure 3: Plots generated by El Agente Cuántico. On the left, potential energy surface (PES) of H2 computed via VQE with the UCCSD ansatz, compared to FCI and HF. On the right, absolute energy errors on a log scale.
• Figure 4: Plots generated by El Agente Cuántico characterizing a 2-qubit Bell state prepared on a quantum simulator. The left and center panels show measurement counts in $Z$ and $X$ bases. The right panel shows the qubit correlations $\braket{Z\otimes Z}$, $\braket{X\otimes X}$, and $\braket{Y\otimes Y}$.
• Figure 5: Plots generated by El Agente Cuántico preparing and validating a thermal state of a 10-site Hubbard chain using imaginary-time evolution.