Research accurate device noise parameters for IonQ, IQM, and Rigetti QPUs

[ddri]

Background

Issue #25 introduced DeviceNoiseModel with pre-configured models for IonQ Aria, IQM Garnet, and Rigetti Ankaa. However, the current parameters are approximate estimates based on general knowledge of each technology type, not verified against actual published specifications.

Current (Unverified) Parameters

Device	T1	T2	1Q Error	2Q Error	1Q Time	2Q Time
IonQ Aria	10s	1s	0.03%	0.5%	10µs	200µs
IQM Garnet	30µs	20µs	0.1%	1%	20ns	60ns
Rigetti Ankaa	20µs	15µs	0.2%	2%	40ns	100ns

Research Tasks

• IonQ Aria: Find official specs from IonQ documentation, papers, or AWS Braket
• IQM Garnet: Find specs from IQM publications, press releases, or partner documentation
• Rigetti Ankaa: Find specs from Rigetti QCS documentation or publications
• Check arXiv: Look for benchmarking papers that characterize these devices
• Check open source: Look for existing noise models in Qiskit, Cirq, or other frameworks

Sources to Check

Vendor Documentation

• IonQ: https://ionq.com/resources, AWS Braket device pages
• IQM: https://www.meetiqm.com/technology, partner integrations
• Rigetti: https://qcs.rigetti.com/qpus, pyQuil documentation

Academic Papers

• arXiv quantum-ph for device characterization papers
• Vendor-authored papers on their hardware

Open Source

• Qiskit Aer noise models
• Cirq device specifications
• Amazon Braket device properties API

Acceptance Criteria

• Update IONQ_ARIA, IQM_GARNET, RIGETTI_ANKAA with verified parameters
• Add source citations as comments in the code
• Document any parameters that vary significantly across publications

Related

• Part of Epic Epic: QuantumFlow Transpilation & Benchmarking Framework #22: QuantumFlow Transpilation & Benchmarking Framework
• Follows Add device noise models for realistic simulation #25: Add device noise models

[ddri]

Research Complete - Parameters Updated

The device noise parameter research has been completed. All three device models now use values from published vendor specifications:

Sources Used

IonQ Aria:

• IonQ Aria System Specifications
• AWS Braket IonQ Aria Launch

IQM Garnet:

• IQM Garnet 20Q Whitepaper 2024
• arXiv:2408.12433

Rigetti Ankaa-3:

• Rigetti Ankaa-3 Press Release
• arXiv:2410.05202

Key Findings

Parameter	Initial Estimate	Researched Value	Device
1Q gate time	10 µs	135 µs	IonQ Aria
2Q gate time	200 µs	600 µs	IonQ Aria
2Q gate error	1.0%	0.49%	IQM Garnet
Readout error	2-3%	5.3%	Rigetti Ankaa

Documentation

• Blog post: docs/blog/2025-01-04-device-noise-models.md
• Technical docs: marqov-chewie docs/technical/quantumflow-circuit-analysis.md

Status

This issue tracks ongoing maintenance of device parameters as vendors update their specs. The initial research is complete, but parameters should be reviewed periodically.

See #28 for future work on dynamic/real-time calibration data integration.