DuoNeural/Sub-Lattice-v3-Lorentz-18M

AI model by DuoNeural

Older snapshot: check the original source for current information.

From the publisher

Model card & documentation

Source preview

Read the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.

Sub-Lattice-v3-Lorentz-18M is the champion and gold medalist of the DuoNeural Geometry Pentathlon, achieving an unprecedented 0.8914 valbpb (14.46 valppl) on held-out validation sequences. By replacing coordinates bounded within the unit ball with the unconstrained Lorentz Hyperboloid manifold in Minkowski spacetime (H² ⊂ ℝ^{256+1}), Lorentz-18M eliminates the boundary denominator singularities of Poincaré and Beltrami-Klein coordinates, establishing a new global benchmark for ultra-compact language models. In the Lorentz model, vectors reside on the upper sheet of a hyperboloid in (d + 1)-dimensional Minkowski space ℝ^{d, 1}: Lorentz-18M exhibits a textbook AdS Holographic Funnel: In contrast, non-hyperbolic baselines (Euclidean and Spherical) exhibit complete Holographic Inversion, where the output layer L02 freezes prematurely at τ > 748k, while L01 fails to crystallize (τ = 41k), leading to catastrophic syntax collapse (0.0% valid JSON). In Sub-Lattice-v3-Lorentz-18M, Crystal Telepathy converged to: Driven by…

DuoNeural/Sub-Lattice-v3-Lorentz-18M on Hugging Face A short preview, not the full document.

Read the full model card ↗ · Preview checked 2026-10-28T20:46:39.949Z

Inside the original model card — Document outline

Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.

Documentation belongs to its respective authors. Reported project/model license: apache-2.0. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.

Model overview

DuoNeural/Sub-Lattice-v3-Lorentz-18M is a task-unspecified model repository published by DuoNeural on Hugging Face. The source reports the pytorch library.

This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.

Model facts

Task: Not reported
Library: pytorch
Recent downloads (30 days): 0
Cumulative likes: 0
Hugging Face trending score: 0
Reported safetensors parameters: 31,601,912
Architecture: Not reported
License: apache-2.0
Access: Not gated by Hugging Face
Created: 2026-10-24T06:29:43.000Z
Last modified: 2026-10-24T06:29:46.000Z

Compatibility and lineage

Reported languages: en

Reported base models: HuggingFaceTB/SmolLM2-360M-Instruct

Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.

Use and evaluate

Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.

No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.

Model card and usage instructions

Model files and configuration

Community discussion

Source and freshness

Source: Hugging Face Hub. Metadata observed 2026-10-24T06:46:27.670Z. Daily imports are snapshots, not real-time monitoring.

Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.

Open original source