advay-platform

Python SDK for the Advay Labs Quantum Platform. Run OpenQASM 3 circuits against our hosted SQPU emulator, get realistic noise projections and surface-code resource estimates, all from a single pip install.

pip install advay-platform

What you get

• Cloud emulator access — submit OpenQASM 3 circuits to dev.advaylabs.com, get counts back.
• Local emulator — the same compute stack also runs offline on your machine for unit tests and CI; no API key required for local use.
• Resource projection — surface-code physical-qubit cost, code distance, T-factory count, and wall-clock for your circuit on the SQPU roadmap device, with factor-of-2 honest bounds.
• Noise projection — end-to-end success probability under three control modes (full / hybrid / reduced), calibrated to the digital twin's gate-time ceilings.
• OpenQASM 3 support — classical control flow (if_else), mid-circuit measurement, and proper feedforward simulation included.
• Qiskit interop — qiskit.QuantumCircuit objects are accepted as input alongside QASM strings.

Quick start

Run a circuit against the cloud API

import os
import requests

QASM = """
OPENQASM 3.0;
include "stdgates.inc";
qubit[2] q;
bit[2]   c;
h  q[0];
cx q[0], q[1];
c[0] = measure q[0];
c[1] = measure q[1];
"""

resp = requests.post(
    "https://advaylabs.com/api/v1/run",
    headers={"Authorization": f"Bearer {os.environ['ADVAY_API_KEY']}"},
    json={"qasm": QASM, "options": {"shots": 1024}},
    timeout=60,
).json()

print(resp["response"]["status"])                 # "succeeded"
print(resp["response"]["ideal_result"]["counts"]) # {'00': 510, '11': 514}

You'll need an API key from the dashboard.

Run a circuit locally (offline)

from advay_platform import run_circuit, RunOptions

result = run_circuit(open("bell.qasm").read(), RunOptions(shots=1024))
print(result.ideal_result.counts)

The full emulator is bundled in the wheel — no network call required. Useful for CI and quick iteration on small circuits.

Typed request options

from advay_platform import RunOptions

opts = RunOptions(
    shots=1024,
    control_mode="hybrid",   # "full" | "hybrid" | "reduced"
    seed=42,                  # repeatable runs
    execute=True,             # set False to skip ideal execution
    resource_estimate=True,
    noise_projection=True,
)

Pass opts.model_dump() as the options field when calling the REST API, or pass the RunOptions object directly to the local run_circuit().

What the response contains

Every run_circuit() call (and every POST /api/v1/run) returns four chunks:

response
├── circuit_summary      # qubits, depth, gate counts, Clifford check
├── ideal_result         # noiseless simulator counts
├── resource_projection  # surface-code physical qubits, code distance,
│                        # T-factories, wall-clock — with factor-of-2 bounds
├── noise_projection     # end-to-end success probability per control mode
└── provenance           # run_id, platform version, twin version, tags

See the response format reference for field-by-field details.

CLI

The wheel installs an advay-platform command:

# Platform info
advay-platform info

# Run a QASM file
advay-platform run bell.qasm --shots 1024 --output-format json

# Pipe QASM from stdin
cat bell.qasm | advay-platform run - --shots 256

Supported gates

Clifford (fast, no qubit cap): i, x, y, z, h, s, sdg, sx, sxdg, cx, cy, cz, swap, iswap, measure, reset, barrier.

Non-Clifford (state-vector simulator, ≤22 qubits by default): t, tdg, rx, ry, rz, p, u, u1, u2, u3, ccx, ccz, cswap, crx, cry, crz, cp, cu, rxx, ryy, rzz, mcx, mcz.

Classical feedback (if_else, mid-circuit measurement, conditional gates) is fully supported in both the local emulator and the cloud API.

Compatibility

Free tier

• 30-day free trial on signup.
• 20 SDK calls per day on the free tier.
• Verified academic emails (.edu, .ac.uk, .ac.in, .edu.au, …) get the free tier with no expiry and no daily cap.

Upgrade to the researcher or commercial tier via support@advaylabs.com.

Documentation

• Getting started: https://advaylabs.com/docs/getting-started
• API reference: https://advaylabs.com/docs/api-reference
• Examples: https://advaylabs.com/docs/examples
• Troubleshooting: https://advaylabs.com/docs/troubleshooting

Support

• Email: support@advaylabs.com
• Status: https://advaylabs.com/status

License

Proprietary. © 2026 Advay Labs Private Limited. All rights reserved.

See LICENSE for terms. Free for evaluation and research use under the preview tier; commercial use requires a license agreement — contact support@advaylabs.com.