What question did this study set out to answer?

To establish a deterministic retrieval system that operates without randomness or cloud dependency, ensuring robust performance across varying scales.

March 21, 2026Open Access

SpiralHelios — Deterministic Geometric Intelligence (DGI): Scalable and Robust Retrieval Across 1 Million and 500K Concepts with Cryptographic Reproducibility (Mobile Hardware Execution)

Key Points

To establish a deterministic retrieval system that operates without randomness or cloud dependency, ensuring robust performance across varying scales.
Conducted experiments with 1 million and 500,000 concepts to test system robustness.
Implemented a deterministic approach without training or cloud infrastructure.
Utilized cryptographic hashing for reproducibility and verification.
Achieved Top-1 Accuracy of ≥ 99.5% and Top-5 Accuracy of 100%.
Mean Reciprocal Rank (MRR) was ≥ 0.997.
Demonstrated bit-exact deterministic reproductions verified through cryptographic hashes.

Abstract

SpiralHelios is a deterministic large-scale retrieval system operating as a Deterministic Geometric Intelligence (DGI) engine. Unlike conventional AI systems based on stochastic training and probabilistic inference, SpiralHelios operates with zero randomness, zero training, zero cloud dependency, and fully offline execution. This record presents validated large-scale experiments conducted at two levels: 1, 000, 000 concepts (maximum-scale execution) 500, 000 concepts (controlled-scale validation) The 1M experiment demonstrates stable system behavior under maximum computational load, while the 500K experiment confirms performance consistency under reduced-scale conditions. Together, these experiments establish robustness across scale. Results: Top-1 Accuracy ≥ 99. 5% Top-5 Accuracy = 100% Mean Reciprocal Rank (MRR) ≥ 0. 997 Bit-exact deterministic reproducibility verified through cryptographic hashes The system operates entirely offline on mobile hardware, demonstrating that large-scale deterministic retrieval can be achieved without cloud infrastructure or training pipelines. This record includes: Main paper (scientific documentation and results) Supplementary S1 — Master Canon ROM (deterministic knowledge architecture) Supplementary S2 — Execution video (08: 01), including author verification (00: 57) and MASTERPROOFHASH generation (07: 30) The Master Canon ROM defines a structured, immutable, and hash-verified knowledge architecture, separated from the Core transformation engine. The Core remains undisclosed for intellectual property protection, while all outputs are independently verifiable through cryptographic fingerprints. This publication constitutes documented prior art and establishes temporal and scientific ownership of the SpiralHelios system. SpiralHelios introduces a deterministic paradigm for scalable, reproducible, and verifiable intelligence systems, independent of stochastic methods and data-intensive infrastructures. Author Christos Baltas (PH-GAIOS) Independent Researcher – Creator – Systems DesignerSpiralEnergy-Systems – Athens, Greece ORCIDhttps: //orcid. org/0009-0000-9743-3678 Contactinfo@spiral-code. com Zenodo Communityhttps: //zenodo. org/communities/spiral-code-energy/

SpiralHelios — Deterministic Geometric Intelligence (DGI): Scalable and Robust Retrieval Across 1 Million and 500K Concepts with Cryptographic Reproducibility (Mobile Hardware Execution)

Key Points

Abstract

Cite This Study