LongCoT-mini - Logic: leaderboard

Long-horizon chain-of-thought reasoning on the easier logic problems of LongCoT-mini (2026), the small companion set of LongCoT, single attempt without tools at the highest available reasoning setting. Accuracy in percent; higher is better.

Metric: Accuracy (%) on the easy logic problems of LongCoT-mini (long-horizon chain-of-thought problems with short inputs and verifiable answers), single-shot, highest available reasoning setting up to the provider output limit, no tools; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 7 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-5.2xHigh53.6#185
2Gemini 3 Pro7.3#145
3Claude Sonnet 4.5Thinking2.7#261
4Kimi K2 (Thinking)2.7
5DeepSeek V3.2Thinking2.7#419
6GLM-4.70.9#380
7Grok 4.1 FastReasoning0.9#427

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=longcot-mini-logic · How It Works · Data refreshed daily, snapshot 2026-11-21.