Software Comparison
Free alternative to Gaussian for small molecules
Serving as an open Gaussian replacement across quantum chemistry workflows is a long-term goal for ChemDFT, not a present reality. Today, ChemDFT provides an accessible browser-based workstation for small-molecule calculations within modest, transparent hosted limits. Here is an honest, objective comparison of current capabilities, strict limits, and future goals.
Sign in with Google. No credit card required. Free for researchers during the beta.
What you can run, and the limits
ChemDFT provides hosted cloud compute for quantum chemistry and materials modeling within strictly defined, transparent boundaries. Calculations execute in isolated Linux worker environments on Amazon Web Services (AWS) pilot hardware, automatically provisioning on demand and shutting down after approximately fifteen minutes of idle time. Because computational resources are shared and free during the beta, hosted system size limits are modest and strictly enforced before submission.
For molecular quantum chemistry using PySCF 2.14.0, the hosted size limit is 12 atoms for geometry optimisation and molecular orbital calculations, and 6 atoms for harmonic vibrational frequencies and time-dependent DFT (TDDFT). For semi-empirical calculations using xTB 6.7.1, the hosted limit is 50 atoms. For conformational search using CREST 3.0.2, the hosted limit is 12 atoms for neutral singlet hydrocarbons and oxygenates. For periodic solid-state calculations using Quantum ESPRESSO 7.6 or periodic PySCF, the unit cell limit is 20 atoms.
ChemDFT does not replace high-performance computing clusters or supercomputers for large macromolecules, heavy polymers, or broad crystal supercells. Every result is provided as a transparent research aid that researchers must independently check and verify before publication. The artificial intelligence assistant plans workflows and explains findings, but it never writes, interpolates, or modifies a numerical calculation result.
While Gaussian supports vast chemical systems up to 250,000 atoms (as documented on gaussian.com Model Chemistries), ChemDFT focuses specifically on small molecules and unit cells within tightly bounded hosted limits: PySCF up to 12 atoms for optimization and orbitals (6 for frequencies/TDDFT), xTB up to 50 atoms, CREST up to 12 atoms (neutral H/C/O singlets only), and Quantum ESPRESSO up to 20 atoms.
Comparing ChemDFT and Gaussian: an honest perspective
Gaussian is a mature commercial computational chemistry suite widely cited in academic and industrial research. Developed over decades, Gaussian offers electronic structure methods ranging from Hartree–Fock and DFT to CASSCF, coupled cluster theory (CCSD(T)), composite thermochemical models (G1–G4, CBS-QB3), and solvent continuum models. However, Gaussian is proprietary commercial software requiring paid institutional licenses, cluster administration, and manual input preparation.
ChemDFT is not an exact clone or a present Gaussian replacement. Becoming an open Gaussian replacement across computational chemistry workflows is our long-term goal, not a present fact. Today, ChemDFT operates as an accessible web workstation for small molecular systems within strictly bounded hosted limits: PySCF up to 12 atoms for optimization and orbitals (6 atoms for frequencies and TDDFT), xTB up to 50 atoms, CREST up to 12 atoms (neutral H/C/O singlets only), and Quantum ESPRESSO up to 20 atoms. ChemDFT tracks immutable cryptographic provenance for every calculation without software licensing fees.
Feature comparison: ChemDFT vs. Gaussian
To evaluate whether ChemDFT meets your computational needs, consider the key architectural and scientific distinctions:
1. Licensing and Access: Gaussian is commercial software licensed per machine, site, or cluster. ChemDFT is free for researchers during the beta, accessible from any web browser without installation, credit cards, or user fees.
2. System Scale: Official Gaussian documentation (gaussian.com Model Chemistries) lists program limits up to 250,000 atoms with linear-scaling methods. In contrast, ChemDFT focuses on small systems within transparent hosted limits: PySCF optimization ≤12 atoms (frequencies/TDDFT ≤6 atoms), xTB ≤50 atoms, CREST ≤12 atoms, and periodic cells ≤20 atoms. Serving as an open Gaussian replacement on large systems remains a long-term goal, not a present fact.
3. File Compatibility: ChemDFT cannot execute Gaussian input files (.gjf or .com) directly. However, ChemDFT incorporates cclib 1.8.1 to parse and import output log files from Gaussian 09, Gaussian 16, and ORCA 5.x–6.1, allowing you to visualize existing results and preserve provenance.
4. Available Methods: ChemDFT provides PySCF 2.14.0 (HF, B3LYP, PBE with STO-3G, 6-31G, def2-SVP, def2-SVPD), GFN2-xTB, CREST conformer searches, and periodic Quantum ESPRESSO 7.6. Advanced methods present in Gaussian—such as coupled cluster (CCSD(T)), MP2, MP4, ONIOM multi-layer models, CASSCF, and specialized solvent reaction fields—are not hosted on ChemDFT.
| Capability / Dimension | ChemDFT (Hosted Beta) | Gaussian (Commercial) |
|---|---|---|
| Access & Licensing | Free during beta; browser-based; no software installation. | Commercial proprietary license; local or cluster installation required. |
| System Size Limits | Small molecules and crystals: PySCF ≤12 atoms, xTB ≤50 atoms, QE ≤20 atoms. | Up to 250,000 atoms documented program limit (gaussian.com Model Chemistries). |
| Electronic Methods | PySCF HF, B3LYP, PBE (gas phase); GFN2-xTB; Quantum ESPRESSO 7.6. | Extensive: HF, DFT, MP2–MP5, Coupled Cluster (CCSD(T)), CASSCF, ONIOM. |
| File Compatibility | Cannot run Gaussian input files; imports Gaussian 09/16 & ORCA output (via cclib). | Native Gaussian input (.gjf, .com) and checkpoint (.chk) execution. |
| Post-HF Methods | Coupled cluster and MP2 are not hosted on ChemDFT. | Full suite of high-accuracy composite methods (G1–G4, CBS-QB3, W1). |
| AI Assistance & Integrity | AI Scientist plans and explains; Result Firewall verifies engine numbers. | Manual scripting or GUI tools (GaussView); no built-in AI Scientist. |
Who should still use Gaussian?
Researchers with advanced electronic structure requirements should continue using Gaussian or dedicated high-performance computing clusters. You will need Gaussian if your research requires:
• High-accuracy benchmark thermochemistry utilizing composite energy methods (CBS-QB3, G3, G4, or W1/W2 theory).
• High-level electron correlation methods such as coupled cluster theory with single, double, and perturbative triple excitations (CCSD(T)).
• Multi-reference systems, transition states requiring complete active space self-consistent field (CASSCF) wavefunctions, or conical intersections.
• Macromolecular complexes, polymers, or explicit solvent cavities exceeding ChemDFT's hosted limits (12 atoms for PySCF or 50 atoms for xTB).
• Complex implicit solvation models (such as SMD, PCM, or COSMO) that are not yet enabled on ChemDFT's hosted workers.
When ChemDFT is the right choice
ChemDFT is ideal when you need rapid, reliable quantum chemistry on small molecules without licensing hurdles. It is well-suited for:
• Chemistry coursework where students explore molecular orbitals, vibrational frequencies, and dipoles without software friction.
• Synthetic chemists checking whether a proposed intermediate has a viable HOMO–LUMO gap or reaction thermochemistry.
• Quick screening of small organic conformers using GFN2-xTB and CREST before intensive cluster calculations.
• Researchers who value cryptographic provenance, immutable execution records, and clean BibTeX citations.
Trademark notice
Gaussian is a registered trademark of Gaussian, Inc. ChemDFT has no affiliation with or endorsement from Gaussian, Inc. References to Gaussian are provided strictly for comparative purposes based on public documentation.
Frequently Asked Questions
Can ChemDFT run my existing Gaussian input (.gjf or .com) files?
No. ChemDFT does not execute Gaussian input files. However, you can import Gaussian 09 and Gaussian 16 output files (.log or .out) into ChemDFT using our cclib-powered parser to visualize geometries, inspect vibrational frequencies, and preserve provenance.
Does ChemDFT support coupled cluster (CCSD(T)) or MP2 calculations?
No. ChemDFT's hosted engines currently support Hartree–Fock, B3LYP, and PBE density functional theory via PySCF, semi-empirical GFN2-xTB, and periodic Quantum ESPRESSO. Post-Hartree–Fock correlation methods are not hosted.
Why are ChemDFT's atom limits smaller than Gaussian's?
ChemDFT provides free cloud computing on dedicated AWS workers for all registered beta users. To keep compute resources fast, predictable, and fair without charging subscription fees, hosted atom limits are restricted to small molecules and unit cells.
How does ChemDFT ensure calculations are authentic without a commercial license?
ChemDFT executes established, peer-reviewed open-source engines including PySCF (BSD-3-Clause), xTB (LGPL-3.0), and Quantum ESPRESSO (GPL-2.0). All calculations run with exact version tracking and SHA-256 cryptographic provenance.