QENEX LAB: A Self-Contained Quantum Chemistry Platform with Sub-Nanohartree Precision and Automated Discovery Capabilities

Read full text Summarise with VIRAT
Authors: Qenex Ai
Cited by: 0
References: 51

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

We present QENEX LAB, an open-architecture quantum chemistry platform implementing Hartree-Fock (RHF/UHF), Møller-Plesset perturbation theory (MP2), coupled cluster singles and doubles (CCSD), the perturbative triples correction CCSD(T), unrestricted CCSD (UCCSD), equation-of-motion CCSD (EOM-CCSD), complete active space self-consistent field (CASSCF), and Kohn-Sham density functional theory (B3LYP, LDA) with linear-response time-dependent DFT (LR-TDDFT). Molecular integrals are evaluated via Obara-Saika recurrence relations with optional libcint acceleration. Benchmark calculations on He, H2, H2O, and other small molecules demonstrate sub-nanohartree agreement with PySCF reference values when using the libcint integral path, and sub-millihartree agreement with the native pure-Python integral engine. We present accuracy benchmarks across STO-3G, cc-pVDZ, and cc-pVTZ basis sets for nine molecules, validate mathematical invariants including Brillouin's theorem, density matrix idempotency, and size consistency, and describe supplementary capabilities including molecular property prediction (ADMET, toxicity, pharmacokinetics), inverse molecular design via genetic algorithms, and a SHA-256 Merkle-tree provenance chain for computational reproducibility. The platform comprises 81 Python source files (90,103 lines) operating without external quantum chemistry dependencies in the computation path.

Fields of Study & Subjects

Advanced Chemical Physics Studies Quantum Computing Algorithms and Architecture Advanced Physical and Chemical Molecular Interactions

Publication details

Journal / source
ChemRxiv
Publication date
26 March 2026
Language
EN
Document type
Preprint