Skip to content

pqid-bench v1.2.1

Release date: 2026-07-31

Version 1.2.1 is a documentation and package-metadata clarification. It does not change the frozen PQID-Bench v1.0.0 scientific object or the behavior of the Python toolkit.

Reader-Facing Ecosystem Map

The package long description and documentation now identify the correct entry point for each common task:

  • PyPI supplies the installable CLI and Python API;
  • Hugging Face supplies the ready-to-use benchmark records and compact core;
  • Zenodo preserves the completed study's full evidence and audit archive;
  • GitHub supplies maintained source, tests, manuals, and development history;
  • GHCR supplies the isolated evaluator container; and
  • OSF preserves the prospective PQID-Bench 2 study contract.

Prospective-Study Boundary

The documentation now explains that an OSF preregistration is an immutable, timestamped plan recorded before the relevant outcomes are collected. PQID-Bench 2 is a separate two-stage future study of semantic validity and controlled quantum-algorithm family contrasts. No PQID-Bench 2 model output or result is part of PQID-Bench v1.0.0.

The current research supplement's O6 label means Overview object 6. It is a cross-reference to the OSF protocol, not another analytical table or current benchmark result.

Version Crosswalk

Dimension Identifier
Python toolkit pqid-bench 1.2.1
frozen benchmark PQID-Bench 1.0.0
evaluator pqid-bench-evaluator-1.1.0-safe-builtins
structural predicate pqid-bench-reference-signature-1.0.0-count-map
evaluator container 1.0.0

The compact core remains byte-identical at SHA-256 d6df084c7acf7a06bc4800f25b952e26f9903ee4a69ce851ab83b7723970c647. No score, model output, target, denominator, or inferential result changed.