Rohan Paul (@rohanpaul_ai) · 10/14/2026, 04:02:34
SanSi Looping Architecture: 1.4B Model Beats 4.2B Baseline on Decision Tasks via Layer Reuse
The SanSi framework enables small pre-trained models like Ouro-1.4B to loop through the same network layers multiple times during inference for complex decision-making. It achieved 72.0% accuracy on 10,027 test decisions, outperforming a non-looped baseline by 13.6 points and surpassing a 4.2B single-pass model on long-chain logic tasks, offering a high-accuracy solution for memory-constrained environments.
SOURCE COVERAGEFull Original Report
Rohan Paul (@rohanpaul_ai) · 2026-10-13 20
This paper finds that small decision models get far more accurate by rerunning the same layers several times before giving an answer.
Many LLM calls only need a choice, like spam or not spam. Typed decision models answer these in a single pass, which fails when a decision needs several dependent steps.
SanSi loops Ouro-1.4B, a model pretrained to reuse its layers, and reads option probabilities after each of 8 loops. Every loop is trained to answer, so the same model can stop anywhere from 1 to 8 loops.
On 10,027 test decisions, SanSi scored 72.0% versus 58.4% for a same-shape model without loops, at 7.7x the compute. On liar chains longer than any seen in training, it reached 72.6% while a 4.2B single-pass model scored 50.0%, the chance level.
If memory limits your decision-style calls, try a looped small model before moving to a model 3x larger.
