'Reasoning depth' estimate flags Sol 6.1 as a suspected looper — but the metric itself may be unreliable
scaling01 · x · 2026-09-30
Using a 'true reasoning depth' estimate, the author ran a spot-the-looper analysis: 5.6 luna/sol and 6 luna/sol show no looping, while 6.1 sol and 6 astra look suspicious. But he concludes the simplest explanation is that the metric isn't very good — and a single loop doesn't double reasoning depth anyway.
Related event: Benchmark sleuths suspect Grok 6.1 of answer-reusing loops(3 posts)→
More from Models
- Stealth Model 'silicon-2' Surfaces in Devin, Suspected to Be Anthropic's Next Model — kimmonismus · 2026-11-16
- Google hides Gemini model label in AI Mode, rolls Gemini 4 Carbon to Pro subscribers — gaganghotra_ · 2026-11-16
- Open-weight Karpo-Decision-31B tops Decision Index at 73.9, one forward pass per answer — CodeByPoonam · 2026-11-16
- Nathan Lambert calls out Ethan Mollick for 'selectively ignoring' Kimi in model comparisons — natolambert · 2026-11-16
- Eval Firm Trajectory Told 40 Agents "NEVER Stop" in Harness Without Telling Epoch — peterwildeford · 2026-11-16
- Unsloth Q8 crushes default Q3 in a self-hosted test over 331 financial PDFs — No_Cartographer3953 · 2026-11-16