Make the workbench
your own.

Scientific analysis and evidence

Understand what happens
under the hood.

ISAES turns retained scientific results into eligible analyses, traceable findings, linked exploration and consistent reports. The dedicated page explains that machinery and the new autonomous investigations: contextual research, permitted computation and claim assessment under an explicit standing policy. The retained walkthrough covers the earlier reporting and reviewed-discussion modes.

Reusable private workflows

From a research question
to Python and a saved workflow.

Learn what the scientific Python program must contain, how to build it from the development guide, and how the workflow assistant can help implement a mathematical specification. Then follow the custom Qiskit study through validation, repeated runs, private versions and B2/B3 controls.

Project access & AI policy

See the controls
behind collaboration.

Watch the B2 and B3 walkthrough for project sharing, viewer and editor grants, revocation, institutional AI policy and disclosure consent. Follow the GUI steps and the corresponding server enforcement checks.

Choose a deployment path

Start where your research works.

Try the live hosted workbench, run PF Simulation locally or deploy your own service. For an introduction, watch the narrated r34 SaaS demo .

Prefer your own environment? Choose standalone use or host a service for your research group.

The hosted trial provides access to the PF Simulation application. Standalone use and your own hosted deployment require the application package; this website provides the supporting documentation and examples.

Guides from the r34 package

From the first question
to the next private layer.

Select a guide to read it here. Each guide includes a table of contents, a formatted download and a print-friendly view.

Connected research examples

Choose a question.
Inspect a complete study.

Start with the integration catalogue to understand applicability and required runtimes. Use the recorded satellite, Aer and ISAES Floquet Gallery bundles to inspect saved inputs, numerical outputs and exact reproduction steps.

Read the methods and numerical scope of the Gallery studies. Read Gallery study validation

Core scientific reference

The wider workbench.

The retained 239-page end-user manual, Edition 1.19.0, covers established workflows, configuration, interpretation and reports. Read the current workbench and maintenance guides alongside it for the October configuration, execution, tracking, cache, workflow and Assistant corrections. Chapter 32 adds an illustrated ISAES walkthrough: user choices and automatic processing, evidence and status interpretation, linked figures, saved views, exports, literature management, optional report discussion and autonomous investigations. The dedicated guides provide additional SDK, AI, BYOK, BYOS and access-control detail.

Cumulative validation across retained releases

Count the coverage.
Keep the evidence in view.

The headline reconciles passing test identities across retained release records from r9 through r34. Repeated runs and copied evidence do not add a new identity. Earlier release totals without surviving identity-level records are not invented. This is cumulative recorded coverage, not a claim that every test ran against the current source.

The operator supplied a successful consolidated HF deployment and startup record on 16 October 2026. That later observation adds no test identities and does not change the frozen package qualification flags. Browser responsiveness, live provider behavior and production capacity require their own evidence.

Acceptance evidence remains tied to its original source and workload. The autonomous and maintenance releases did not rerun the full scientific benchmark campaign. Website checks remain separate from the application totals. Native Windows, representative researcher usability, applicable production/BYOS deployment and live-model scientific accuracy remain separate qualification gates.