The Edge-Native Quantum OS for Utility-Scale Applications

Bypass the Hardware. Resolve the Impossible.

Catalyst-Q is the world's first stateless quantum execution proxy. We have replaced localized cryogenic hardware with a software-defined topological architecture. Stop waiting a decade for fault-tolerant hardware—execute massive 2,500-asset risk models and 256-qubit material simulations natively on the edge today.

Run live API benchmarks Review proof Start building

pip install catalyst-q

# Controlled hosted index:
pip install --index-url https://catalyst-q-sdk.strategic-innovations.ai/simple catalyst-q

Public benchmark harness

Catalyst-Q ships a reproducible public benchmark harness in catalyst_q.benchmarks. It prepares SDK requests against respected public benchmark families and publishes a downloadable JSON artifact. The harness validates request generation, payload shape, billing estimates, and suite coverage. It is not a formal complexity proof and does not describe private execution internals.

Download JSON artifact Download live demo artifact Download high-qubit evidence Download SATLIB evidence Install SDK

Benchmark families 9 suites

QED-C, SupermarQ, QASMBench, MQT Bench, SATLIB, TSPLIB, OR-Library, and Biq Mac solver datasets.

Runnable cases 11 cases

Five quantum request-prep cases and six solver request-prep cases.

Free-tier fit 16 max qubits

All bundled cases fit inside the public 50-qubit developer tier.

Official corpus evidence

The current published official-corpus run covers SATLIB uf20-91: 1,000 named SATLIB random 3-SAT instances with 3,000 raw result rows across Catalyst-Q live API, Kissat, and CaDiCaL. It validates hosted SAT correctness on this corpus; it is not a SAT runtime superiority claim.

Download summary JSON Download report Download raw JSONL

Solver	Valid records	Valid rate	Median runtime seconds	Total runtime seconds
Catalyst-Q live API	1000 / 1000	1.000000	0.54528125	748.303935
Kissat local baseline	1000 / 1000	1.000000	0.011456708	10.898355
CaDiCaL local baseline	1000 / 1000	1.000000	0.009249604	8.845314

Source SHA256: be2835295e8500bb28f0314eba70bd0deaff1250b187260f7b6d0772bdf111a5. Source: https://www.cs.ubc.ca/~hoos/SATLIB/Benchmarks/SAT/RND3SAT/uf20-91.tar.gz.

The second published run covers OR-Library mknap1: 5 of 7 official multidimensional knapsack instances, with Catalyst-Q live API matching the official optimum on every selected case.

Download OR-Library summary Download OR-Library report Download OR-Library raw JSONL

Solver	Valid records	Valid rate	Median runtime seconds	Total runtime seconds
Catalyst-Q live API	5 / 5	1.000000	0.076966708	0.378504
Exact B&B reference	5 / 5	1.000000	0.002967	0.014968

Source SHA256: 727c5f90b6acafa0896ce4b5b5559e2995303b735ee083a07e9b724738fac283. Source: https://people.brunel.ac.uk/~mastjjb/jeb/orlib/files/mknap1.txt.

Run it locally

catalyst-q-benchmark --output-dir catalyst-q-public-benchmarks

catalyst-q-benchmark --execute-api --output-dir catalyst-q-live-api-benchmarks

from catalyst_q import write_public_benchmark_artifacts

write_public_benchmark_artifacts("public-benchmark-results")
write_public_benchmark_artifacts("live-results", execute_api=True, timeout=30)

cd catalyst-q-benchmarks
PYTHONPATH=src:../sdk/python python scripts/run_official_corpus.py \
  --corpus satlib-uf20 \
  --execute-api \
  --policy-tier enterprise \
  --output-dir results/official_corpora/satlib_uf20_91

The generated report includes per-case suite, family, workload type, size label, compute units, payload bytes, request route, and citation URL. With --execute-api, each row also records status_code, latency_ms, response_bytes, and response_sha256.

Rigor guardrails

Use fixed deterministic cases for repeatable SDK results.
Keep public claims tied to generated artifacts.
Separate request-preparation validation from hosted execution results.
Use public suite names and source URLs instead of internal method names.

Coverage chart

QED-C BV

285,696 CU

SupermarQ GHZ

524,288 CU

SupermarQ QAOA

393,216 CU

QASMBench QFT

112,640 CU

MQT Grover

282,624 CU

SATLIB 3-SAT

1,720 CU

TSPLIB TSP

100 CU

OR-Library KP

576 CU

OR-Library portfolio

64 CU

Biq Mac QUBO

36 CU

Biq Mac Max-Cut

54 CU

Case matrix

SECURITY DISCLAIMER: 4099-Qubit Shor's Surrogate

The shor_rsa_2048_surrogate benchmark is purely a surrogate scaling proof used to validate the SDK's ability to compress and transmit the $2^{4099}$ amplitude space of a massive Modular Multiplier block. It is NOT a full factorization of an RSA-2048 key. Catalyst-Q does not condone or support the unauthorized breaking of public key infrastructure.

ID	Suite	Family	Type	Size	Compute units	Payload bytes
`qedc_bernstein_vazirani_8`	QED-C Application-Oriented Benchmarks	Bernstein-Vazirani	circuit	8-bit hidden string	285,696	1,778
`supermarq_ghz_16`	SupermarQ	GHZ	circuit	16 qubits	524,288	2,060
`supermarq_qaoa_ring_8`	SupermarQ	QAOA ring	circuit	8 qubits, one layer	393,216	2,781
`qasmbench_qft_5_qasm`	QASMBench	QFT	qasm	5 qubits	112,640	1,430
`mqtbench_grover_6`	MQT Bench	Grover	circuit	6 qubits	282,624	2,107
`satlib_random_3sat_20`	SATLIB	Random 3-SAT	solver	20 variables, 86 clauses	1,720	1,737
`tsplib_euclidean_10`	TSPLIB	Euclidean TSP	solver	10 cities	100	1,166
`orlib_mknap_12`	OR-Library Knapsack	0/1 knapsack	solver	12 items	576	883
`orlib_portfolio_8`	OR-Library Portfolio	Cardinality-constrained portfolio	solver	8 assets	64	1,432
`biqmac_qubo_6`	Biq Mac Library	Binary quadratic optimization	solver	6 binary variables	36	985
`biqmac_maxcut_6`	Biq Mac Library	Max-Cut	solver	6 nodes, 9 weighted edges	54	1,049
`cafa6_dag_optimization_12`	Catalyst Biocomputation	DAG Optimization	solver	12 nodes, 11 dependencies	144	1,494
`cafa6_full_adder_4`	Catalyst Arithmetic	Adder	circuit	4 qubits	36,864	1,197
`google_echo_8`	Catalyst Sampling	Loschmidt Echo	circuit	8 qubits	442,368	2,651
`sat_competition_100`	SAT Competition	Random 3-SAT	solver	100 variables, 400 clauses	5,732	1,578
`miplib_mknap_100`	MIPLIB 2017	Integer Linear Programming	solver	100 items, 5 constraints	2,427	564
`qasmbench_qft_500`	QASMBench Advantage	Quantum Fourier Transform	circuit	500 qubits, 125,250 gates	8,903,542	173,722
`benchpress_tfim_100`	SupermarQ & IBM Benchpress	Hamiltonian Simulation	circuit	100 qubits, 10 steps	186,960	13,429
`supermarq_qaoa_10000`	SupermarQ & IBM Benchpress	QAOA	circuit	10,000 qubits, 285,000 gates	14,038,141	209,077
`qchem_femoco_108`	Quantum Chemistry (VQE)	Hamiltonian Simulation	circuit	108 qubits, 50,000 gates	2,068,774	49,968
`shor_rsa_2048_surrogate`	Cryptanalysis (Shor's)	Modular Exponentiation	circuit	4,099 qubits, 100,000 gates	4,262,951	84,547
`darpa_qbi_logistics_vrp_1000`	DARPA QBI (Utility Scale)	Vehicle Routing	solver	1,000 nodes, 100 vehicles	100,000,000	5,977,562
`darpa_qbi_power_grid_2000`	DARPA QBI (Utility Scale)	Unit Commitment	solver	2,000 generators, 24 periods	1,152,000	35,067
`darpa_qbi_finance_2500`	DARPA QBI (Utility Scale)	Portfolio Optimization	solver	2,500 assets	6,250,000	37,532,593
`darpa_qbi_materials_hubbard_256`	DARPA QBI (Utility Scale)	Materials Science	circuit	256 qubits, 40,000 gates	19,260,243,968	2,910,233

External benchmark sources

Suite	Why it matters	Source
QED-C Application-Oriented Benchmarks	Application-oriented quantum benchmarks with volumetric performance framing.	QED-C publication
SupermarQ	Scalable application benchmarks and hardware-agnostic feature vectors.	SupermarQ paper page
QASMBench	OpenQASM suite for NISQ evaluation, compilers, schedulers, and simulators.	PNNL QASMBench
MQT Bench	Cross-level benchmark generation for quantum software tooling.	MQT Bench
SATLIB	Uniform SAT benchmark test bed using DIMACS CNF conventions.	SATLIB
TSPLIB	Standard TSP and related routing problem instance library.	TSPLIB
OR-Library	Operations research datasets for knapsack and portfolio optimization.	Knapsack / Portfolio
Biq Mac Library	Binary quadratic optimization and Max-Cut instances for optimization method testing.	Biq Mac Library