Opus 5 aces 20-digit multiplication; speculation on why no-thinking modes are withheld
scaling01 · x · 2026-09-17
A simple eval by maksymandr shows Claude Opus 5 hits 100% accuracy (1200/1200) on multiplications up to 20x20 digits, but only at max reasoning—a useful probe for no-CoT capability, especially for recurrent-depth architectures like GPT-6 Astra. scaling01 speculates Fable and Astra lack a no-thinking mode because it would leak hints about model size and depth, widen the attack surface, and risk a weak non-CoT Fable breaking the hidden-Fable-CoT mechanism.
Related event: Frontier LLMs Nail 20-Digit Multiplication Without Tools(9 posts)→
More from Models
- 'You Can Feel the Gradient Slopes': Dev Contrasts Benchmaxxed RL Models vs General Ones — theteknosaur · 2026-11-02
- LightOnOCR-3 Launches: Tops FRBench, Adds Layout Structure and Chart Extraction — IgorCarron · 2026-11-02
- NVIDIA's Kumo Is a Family of Four Tabular Foundation Models, Not Just One — pandeyparul · 2026-11-02
- SemiAnalysis: safety classifiers are underrated — Meta's Muse tops cyber benchmark at 35.2% — ThomasScialom · 2026-11-02
- DeepSeek V4.1 Flash 1017 wins over a longtime closed-source-only user: fast, cheap, reliable — _bachrc · 2026-11-02
- Local quants vs Claude on 9 real private bugs: 3-bit costs thinking time, not score — FeydRowan · 2026-11-02