Third-party validated (TestFort QA Lab, Dec 2025)

Deterministic vector math at GPU scale

Bit-exact evaluation of dense numeric expressions (y: f(x)) via a stateless API. SHA-256 verification supported.

Memory-safe Rust core | SIMD + GPU acceleration | Edge to enterprise

286.94B
ops/sec
444.4T
ops (1-hr test)
0%
error rate
2.35B
ops/joule

"The Only Engine That Guarantees Reproducibility"

NumPy, PyTorch, and TensorFlow optimize for speed. They do not guarantee identical results across runs or platforms. GPU thread scheduling causes floating-point drift that breaks audits, blocks certifications, and makes debugging impossible.

Lu(x)i is the only math engine that guarantees bit-exact, SHA-256 verifiable results on ARM, x86, and GPU, every time.

Why Efficiency Matters

"Faster Math. Less Power. Less Heat."

Lu(x)i completes vector math workloads faster than standard libraries. Faster completion means less time at peak power.

Edge/Battery Systems

Extended runtime, reduced thermal throttling

Data Centers

Less heat per workload, lower cooling demand

Validated: 2.35 billion ops/joule
NVIDIA H100 (TestFort, Dec 2025)
See the Thermal Simulator →

"Efficiency validated for dense vector expressions (y = f(x)). Results vary by workload and hardware."

Validated by

QA Lab
PFLB
Load Testing
FPBench
Trust Bar
GitHub
[Verified]

Dec 2025

Defense & Autonomous Systems

Deterministic vector math for safety-critical edge compute

DO-178C certification-ready architecture with SHA-256 verified calculations for targeting systems, sensor fusion, and autonomous navigation.

  • DO-178C certification-ready architecture
  • Sub-5MB container, Jetson-ready
  • 2.35B ops/joule for battery-powered systems
  • SHA-256 verified targeting calculations
  • Cross-platform determinism (ARM, x86, RISC-V, WASM)
  • Bit-exact reproducibility for certification

Quantitative Finance

FINRA-compliant reproducible calculations

Reproducible calculations for trading, risk modeling, and Monte Carlo simulations with complete regulatory audit trail support.

  • FINRA Rule 3110 compliant reproducibility
  • Bit-exact Monte Carlo simulations
  • Reproducible backtesting and VAR models
  • Audit-ready SHA-256 verification
  • Same input = same output, always
  • Regulatory audit trail support

Every Watt Counts

Faster compute means less time at full power

Lu(x)i completes calculations faster, so your hardware spends less time at full power.

  • Edge & Battery Systems: Drones, robots, satellites run longer between charges
  • Data Center Savings: Faster compute = less heat = reduced cooling costs
  • 2.35B ops/joule: Validated efficiency on NVIDIA H100
  • Scale savings: Facilities running Lu(x)i save enough electricity annually to power hundreds of homes

Validated Performance Metrics

Independently verified by TestFort QA Lab during a 1-hour GPU endurance test on NVIDIA H100 SXM, December 2025.

Reliability & Compliance

Metric Value Why It Matters
Determinism SHA-256 verified Bit-exact results for audits & certification
Error Rate 0.00% Zero failures over 444.4T operations
Hash Consistency 10/10 runs identical GPU and CPU produce same output

Performance

Metric Value Why It Matters
Aggregate Throughput 286.94B ops/sec Process massive vectors in milliseconds
Peak Throughput (sqrt) 331.13B ops/sec Function-dependent ceiling
P95 API Latency 1.47ms Real-time capable under load

Efficiency

Metric Value Why It Matters
Energy Efficiency 2.35B ops/joule Edge-deployable, battery-friendly
Average GPU Power 117.2W Below H100 TDP

Test Conditions

Parameter Value
Hardware NVIDIA H100 SXM, 80GB HBM3
Duration 1 hour continuous
Concurrent Users 200
Total Operations 444.4 trillion
Validator TestFort QA Lab, Dec 2025

Verified Output Hash (10 consecutive runs)

98bd97026a738671ec7c3d302efa6aa8ff078a5fb9183f7fdf51a1c4ff938321

Identical hash confirms bit-exact determinism across GPU and CPU execution modes.

Verified Identity

Verified: Cross-Platform Determinism

Zero drift across incongruent hardware architectures via the 'Art of Fugue' Polyphonic Stress Test.

Platform Architecture Stress Test SHA-256 Identity
Apple M1 Pro ARM64 (Neon) Polyphonic (40 Threads) 98bd97...ac19 [MATCH]
NVIDIA H100 CUDA (Ampere) Polyphonic (40 Threads) 98bd97...ac19 [MATCH]
NVIDIA L4 Vulkan Polyphonic (40 Threads) 98bd97...ac19 [MATCH]

Core Capabilities

Deterministic Execution

Same input + same environment = identical output with SHA-256 verification.

Dual Execution Modes

GPU mode (CUDA/Vulkan FP16/FP32) and CPU mode (Rust f32 with SIMD).

High Throughput

Demonstrated 286.94B ops/sec aggregate on NVIDIA H100 SXM.

Energy Efficient

2.35B ops/joule efficiency enables edge and power-constrained deployments.

REST API

Stateless HTTP interface for deterministic expression evaluation.

POST /evaluate

Vectorized expression evaluation with deterministic results.

Request:

POST /evaluate
{
  "expr": "2*x + cos(x)",
  "x": [1.0, 2.0, 3.0]
}

Response:

{
  "y": [2.5403, 4.5839, 6.0100]
}

expr = mathematical expression | x = scalar or vector input | Deterministic: same request = same response

For Researchers & Academics

Transparent methodology, reproducible benchmarks, and citation formats for academic evaluation.

For Researchers

  • 01 SHA-256 verification for bit-exact reproducibility
  • 02 Documented methodology with hardware specifications
  • 03 Third-party validation by independent QA lab
  • 04 Citation formats available for academic papers

Methodology

  • Polyphonic Verification: Validated using the 'Art of Fugue' suite, simulating 40 concurrent threads of trigonometric, logarithmic, and transcendental variance to force scheduler drift.
  • GPU mode uses IEEE 754 FP16/FP32 with defined rounding
  • CPU mode uses Rust native f32 with SIMD (AVX-512/AVX2/ARM Neon)
  • Throughput = total floating-point operations per second

BibTeX

@software{luxiedge2025,
  author       = {Waller, Eric},
  title        = {{Lu(x)i}: Deterministic JSON 
                  Math Engine},
  year         = {2025},
  version      = {1.0},
  url          = {https://luxiedge.com},
  note         = {Accessed: 2025-12-18}
}

APA Style

Waller, E. (2025). Lu(x)i: Deterministic JSON Math Engine (Version 1.0) [Computer software]. https://luxiedge.com

Note: Lu(x)i is commercial software. This citation references the software directly.