qb-compiler listed on IBM Quantum Ecosystem · Error Mitigation + Circuit Optimization

THE PERFORMANCE LAYER FOR QUANTUM ERROR CORRECTION

QubitBoost lowers your logical error rate by correcting the error model your decoder reads, before it decodes, without replacing it. Decoding performance is in private pilot. Around it sits a hardware-validated platform you can run today that improves runtime efficiency and result assurance.

Every number on this page traces to a hardware experiment directory. Every claim is reproducible. No synthetic benchmarks, no cherry-picked baselines.

DECODING PERFORMANCE

lower logical error rate, any decoder

DECODER ADAPTOR

private pilot

We correct the error model your decoder reads, before it decodes, so the decoder you already run reports a lower logical error rate. No decoder swap, no retraining, with a built-in guard so it is never made worse.

Request pilot access

2 platforms

Validated on real neutral-atom and superconducting hardware data

MWPM + BP+OSD

Improves the decoders teams already run

No swap

Slots in front of the decoder you already have

Never worse

Built-in no-harm guard

RUNTIME EFFICIENCY

fewer shots and evaluations for the same answer

VQE Cost Reduction

ChemGate

Lower-cost adaptive VQE.

44.2% eval reduction · IBM Fez caffeine 16q (n=8 hw)

Reaches HF energy on n=8 replicates (Apr 25 2026)
26-28% across 5 drugs in simulator (metformin, acetaminophen, omeprazole)
Hardware validated Apr 2026

Shot Validation

ShotValidator

Flags and drops corrupted QEC shots.

74% mean capture · 6/9 IBM Fez configs hit 100%

One config (d=5/r=8/f=3) is net-negative -12.2%, disclosed in claims doc
216K shots validated on IBM Fez
Flags and drops corrupted QEC shots

RESULT ASSURANCE

know when to trust a run, and prove it afterward

Hardware Observatory

DriftRadar

Continuous multi-backend drift tracking for routing, abort, and post-hoc audit.

r = 0.9996 drift-score vs odd-parity readout-event rate on IBM Kingston

253K hardware shots validated
6/6 gates pass · FPR 0%
20s latency on 10K-shot checks

Real-Time Abort

LiveGate

Abort runs at round 2 when the signal says they will not recover.

AUC 0.654 round-2 abort · IBM Fez 50K shots

21.8% lift vs random baseline
80% budget saved on aborted runs
Early rounds beat late (hw-only insight)

QEC Postselection

SafetyGate

In-flight trust gating for decoded QEC shots.

Protected-logical: ~14% fewer logical errors at 20% discard on a degraded-day Fez run (single-job), plus 38.8% on IBM Torino via postselection

Reproduced on real Google Willow: +26 / +35 / +53% LER cut at d = 3 / 5 / 7
Block-stratified, 5-fold cross-validated
Drift-stratified weight-only baseline: ~7.7%
Cross-backend replication on Torino + Ankaa-3 in queue

Compliance Audit

QC-Attest

Blockchain-anchored Result Passports for regulated environments.

Merkle + Polygon + HMAC · 6.1ms local sign

Tamper-evident by construction
GxP, SOX, GDPR quantum workflows
Per-epoch root commit to Polygon PoS

Certified Entropy

QRNG

NIST SP 800-22 certified quantum randomness as a service.

IBM Fez production, von Neumann extracted, signed batches

Von Neumann extracted
EUR 3.20 / MB · IBM Fez production
Signed HMAC provenance certificate per batch

READY TO EVALUATE?

2-week pilot on your workloads. No commitment. No credit card.

SEE IT LIVE REQUEST PILOT ACCESS