[NEW PASS] Modular compilation for Chiplet architectures

[jordandsullivan]

Short description

Technique name: Modular compilation for Chiplet architectures

This paper actually describes a variety of different compiler passes, each of which could be implemented separately and eligible for its own #merit-bounty.

...the complexity of chiplet-based quantum devices, coupled with their growing size,
presents an imminent scalability challenge for quantum compila-
tion. Contemporary compilation methods are not well-suited to
chiplet architectures — in particular, existing qubit allocation meth-
ods are often unable to contend with inter-chiplet links, which
don’t necessarily support a universal basis gate set. Furthermore,
existing methods of logical-to-physical qubit placement, swap inser-
tion (routing), unitary synthesis, and/or optimization, are typically
not designed for qubit links of significantly varying latency or fi-
delity. In this work, we propose SEQC, a hierarchical parallelized
compilation pipeline optimized for chiplet-based quantum systems,
including several novel methods for qubit placement, qubit routing,
and circuit optimization. SEQC attains a 9.3% average increase in
circuit fidelity (up to 49.99%). Additionally, owing to its ability to
parallelize compilation, SEQC achieves 3.27× faster compilation on
average (up to 6.74×) over a chiplet-unaware Qiskit baseline.

Source of technique

Scientific/white-paper: https://arxiv.org/abs/2501.08478
Code repository: Need to ask authors

---

Full proposal (optional)

If you want to actively work on integrating this pass, please fill out our formal
New Compiler Pass Proposal discussion template and link it here:

[mrmoe1985-sketch]

I'd like to work on this bounty. I've already submitted PRs #641 (QMAP routing) and #642 (Discord bot).

I can implement the SEQC pipeline for chiplet architectures. The key components would be:

1. Chiplet topology representation
2. Hierarchical qubit placement
3. Inter-chiplet routing with latency-aware swaps
4. Parallelized compilation

Let me know if you'd like me to start with a specific component first. Looking forward to working on this!

[mrmoe1985-sketch]

Hi! I'd like to work on this bounty.

I notice this issue is from September 2025. Is it still active and eligible for the Merit bounty payment?

I have experience with chiplet architectures and can implement:

• Chiplet topology representation
• Hierarchical qubit placement
• Inter-chiplet routing with latency-aware swaps

Could you confirm if this is still open for contribution? Thanks!