API Documentation

OpenQP exposes several layers of programmatic access. Most users should start with the Python API because it covers both compact OpenQP scripts and direct Runner execution of existing input files.

API Layers

Layer	Entry point	Audience	Stability
Command line	openqp input.inp	Production runs, shell scripts, workflows	Stable
Python API	oqp.openqp.OpenQP, oqp.pyoqp.Runner	Scripts, notebooks, services, and automated workflows	Recommended
Molecule data	runner.mol	Result extraction and advanced workflows	Semi-stable
Input checker	oqp.utils.input_checker.check_input_values	Input builders, web forms, preflight validation	Recommended
Convenience wrapper	oqp.openqp.OPENQP	Compact dotted-key Python inputs	Experimental
C/Fortran handle	include/oqp.h, oqp_handle_t	OpenQP developers and native extensions	Internal

The import package is oqp, even though the source tree contains a pyoqp directory and the installable distribution is named OpenQP.

Recommended Starting Points

• Use Run OpenQP from Python for complete script-based calculation examples.
• Use Python API for high-level OpenQP scripts, direct Runner execution, and the supported Python helper functions.
• Use Results and Molecule Data to retrieve energies, gradients, orbitals, TDDFT arrays, SOC data, and serialized result dictionaries.
• Use Input Validation when building forms, agents, or custom front ends that need actionable diagnostics before launching OpenQP.
• Use the Developer Guide when contributing new keywords, workflows, native kernels, data tags, or bindings.

Generated References

The manual pages above are the curated API guide. The older generated references remain useful for source browsing:

• Generated Python API reference
• Generated Fortran API reference

Generated references can expose internal modules that are not intended as stable user-facing APIs. Prefer the manual pages for supported scripting patterns.