v1 — Qwen2.5 cross-scale + Qwen3-32B headline crossover (the original v1 intent per the 2026-04-11 writeup). 5 model repos on HuggingFace; see john-broadway.

v2 Qwen3-family cohort (this card's cohort — parallel Qwen3-family RYS-applied weights, April 2026):

0.6B — john-broadway/Qwen3-0.6B-RYS-10-13-GGUF
1.7B (this card) — john-broadway/Qwen3-1.7B-RYS-7-10-GGUF
8B — john-broadway/Qwen3-8B-RYS-16-19-GGUF

v2 cross-architecture corpus (21 model variants spanning 10 architecture families): john-broadway/rys-sovereign-collection-v2

Attribution: John Broadway, with collaboration from Claude (Opus 4.6 in April 2026 build; Opus 4.7 in May 2026 cross-architecture analysis and family-relabeling). Original RYS method by David Ng on Qwen2-72B; sweep toolkit by alainnothere.

v2 cross-architecture context (2026-05-13)

This model's place in the v2 curve: baseline reasoning 82.35%, peak RYS Δ +11.76%. The (7,10) configuration is in the early-mid layer band typical of 28-layer architectures.

Across the 21 model variants (10 architecture families) surveyed in john-broadway/rys-sovereign-collection-v2:

Pearson r(baseline reasoning, peak RYS lift) = −0.726. Weak baselines lift more, in their weakest dimension.
Three RYS-recoverable suppression mechanisms identified: under-training scale, MoE routing inefficiency, specialization training trade-off.
One published negative result (SmolLM2-1.7B). RYS is not universal.

v2 attribution: John Broadway, with cross-architecture analysis by Claude (Opus 4.7). Original RYS method by David Ng; circuit-finder toolkit by alainnothere.

john-broadway/Qwen3-1.7B-RYS-7-10-GGUF

Qwen3-1.7B-RYS-7-10

Results

Usage

Full sweep data

Where this sits in the Sovereign Collection

v2 cross-architecture context (2026-05-13)

john-broadway/
Qwen3-1.7B-RYS-7-10-GGUF