Official Resources
- Homepage: https://quantumpackage.github.io/qp2/
- Documentation: https://quantumpackage.github.io/qp2/
- Source Repository: https://github.com/QuantumPackage/qp2
- License: GNU Affero General Public License v3.0
Overview
Quantum Package is a programming environment for quantum chemistry developed in France, focusing on wavefunction methods and quantum Monte Carlo. Developed primarily at the Laboratoire de Chimie et Physique Quantiques in Toulouse, Quantum Package provides a modular framework for implementing and developing wavefunction-based methods with emphasis on selected configuration interaction and stochastic approaches.
Scientific domain: Wavefunction methods, selected CI, quantum Monte Carlo, method development
Target user community: Method developers, wavefunction theory researchers, QMC specialists
Theoretical Methods
- Selected configuration interaction (SCI)
- CIPSI (Configuration Interaction using a Perturbative Selection made Iteratively)
- Full configuration interaction (FCI)
- Quantum Monte Carlo (VMC, DMC)
- Hartree-Fock
- Density Functional Theory
- Coupled cluster
- Multi-reference methods
- Stochastic approaches
- Determinant-based methods
Capabilities (CRITICAL)
- Ground-state electronic structure
- Selected CI calculations
- CIPSI method
- Full CI for small systems
- Quantum Monte Carlo
- Excited states
- Multi-reference wavefunctions
- Benchmark-quality accuracy
- Stochastic methods
- Method development platform
- Modular architecture
- Python interface
- Parallel execution
- Research and development tool
Sources: GitHub repository (https://github.com/QuantumPackage/qp2)
Key Strengths
Selected CI:
- CIPSI algorithm
- Efficient wavefunction compression
- Controlled accuracy
- Large active spaces
- Systematic convergence
Modularity:
- Plugin architecture
- Easy method development
- Extensible framework
- Research platform
- Custom modules
QMC:
- Variational Monte Carlo
- Diffusion Monte Carlo
- Stochastic approaches
- High accuracy
- Benchmark quality
Open Source:
- AGPL v3 licensed
- GitHub repository
- Free to use
- Community development
- Transparent code
Method Development:
- Research tool
- Algorithm testing
- New method implementation
- Flexible framework
- Educational value
Inputs & Outputs
-
Input formats:
- EZFIO database format
- XYZ coordinates
- Basis set specifications
- Python interface
-
Output data types:
- Energies and wavefunctions
- CI coefficients
- Molecular orbitals
- QMC data
- Analysis files
Interfaces & Ecosystem
-
Python Interface:
- Python API
- Scripting capabilities
- Workflow automation
- Custom analysis
-
EZFIO:
- Hierarchical database
- Data storage
- Input/output management
- Efficient access
-
Development:
- Plugin system
- Module development
- GitHub collaboration
- Community contributions
Workflow and Usage
Typical Workflow:
- Create EZFIO database
- Run SCF calculation
- Select CI method (CIPSI)
- Run wavefunction calculation
- Analyze results
- Optional: QMC for refinement
Command-Line Usage:
# Create database
qp_create_ezfio molecule.xyz -b cc-pvdz
# Run Hartree-Fock
qp_run scf molecule.ezfio
# Run CIPSI
qp_run fci molecule.ezfio
Python Interface:
from qp import *
# Python scripting for workflows
Advanced Features
CIPSI:
- Iterative selection
- Perturbative correction
- Systematic improvement
- Efficient for large systems
- Benchmark quality
Stochastic Methods:
- Monte Carlo sampling
- Stochastic CI
- Variance reduction
- Controlled accuracy
- Scalable
Multi-Reference:
- Large active spaces
- Multiple configurations
- Strongly correlated systems
- Accurate description
- Flexible selection
Plugin System:
- Custom modules
- Method development
- Research extensions
- Community plugins
- Easy integration
Performance Characteristics
- Speed: Varies by method
- Accuracy: Benchmark quality for selected methods
- System size: Small to medium molecules
- Scalability: Good parallelization
- Typical: Research calculations
Computational Cost
- CIPSI: Expensive but efficient
- FCI: Very expensive (small systems)
- QMC: Stochastic, time-dependent
- Development focus: Not production speed
- Research: Accuracy-focused
Limitations & Known Constraints
- Speed: Not optimized for production
- Documentation: Research-level
- Learning curve: Steep
- Community: Smaller, specialized
- Focus: Method development vs production
- Platform: Linux primarily
- Maturity: Research tool
Comparison with Other Codes
- vs Production codes: QP development-focused
- vs MOLPRO/MOLCAS: QP more modular, research-oriented
- vs GAMESS/NWChem: QP specialized for CI/QMC development
- Unique strength: Selected CI (CIPSI), modularity, QMC, method development platform
Application Areas
Method Development:
- Algorithm research
- New methods
- Testing approaches
- Benchmarking
- Code prototyping
Benchmark Calculations:
- Reference energies
- Accuracy standards
- Method validation
- Small system exactness
- Quality assessment
Strongly Correlated:
- Multi-reference systems
- Bond breaking
- Transition metals
- Selected CI advantages
- Accurate treatment
QMC Studies:
- Quantum Monte Carlo
- Stochastic methods
- High-accuracy energies
- Research applications
Best Practices
Method Selection:
- CIPSI for general use
- FCI for small systems
- QMC for refinement
- Appropriate for system
- Systematic approach
Convergence:
- CIPSI threshold
- Selection criteria
- PT2 energy
- Systematic improvement
- Check convergence
Development:
- Use plugin system
- Follow architecture
- Contribute to GitHub
- Test thoroughly
- Document code
Community and Support
- Open-source (AGPL v3)
- GitHub repository
- French research community
- User mailing list
- Active development
- Research collaborations
Educational Resources
- GitHub documentation
- Wiki pages
- Example calculations
- Published papers
- Tutorials (growing)
Development
- Laboratoire de Chimie et Physique Quantiques (Toulouse)
- Anthony Scemama
- Michel Caffarel
- French research groups
- International collaborations
- Active GitHub development
Research Focus
CIPSI Method:
- Selected configuration interaction
- Perturbative selection
- Iterative improvement
- Benchmark quality
- Research applications
Stochastic Approaches:
- Monte Carlo methods
- Variance reduction
- Scalable algorithms
- High accuracy
- Active research
Modularity:
- Framework for development
- Easy prototyping
- Method testing
- Community contributions
- Extensible design
French Development
- Strong French tradition
- Toulouse expertise
- European collaboration
- Academic excellence
- Research-driven
Technical Innovation
Selected CI:
- Efficient wavefunction representation
- Systematic selection
- Perturbative corrections
- Scalable approach
- Benchmark quality
Modular Architecture:
- Plugin-based
- EZFIO database
- Clean separation
- Easy extension
- Research platform
Verification & Sources
Primary sources:
- Homepage: https://quantumpackage.github.io/qp2/
- GitHub: https://github.com/QuantumPackage/qp2
- A. Scemama et al., J. Chem. Theory Comput. papers on Quantum Package
- CIPSI method publications
Secondary sources:
- GitHub documentation
- Published studies using Quantum Package
- Selected CI literature
- French quantum chemistry community
Confidence: LOW_CONF - Research tool, specialized methods, smaller community
Verification status: ✅ VERIFIED
- GitHub: ACCESSIBLE
- Documentation: Available online
- Source code: OPEN (GitHub, AGPL v3)
- Community support: Mailing list, GitHub
- Academic citations: Growing
- Active development: Regular GitHub activity
- Specialized strength: Selected CI (CIPSI), quantum Monte Carlo, modular architecture, method development platform, wavefunction methods, research tool, stochastic approaches