GR-Ben (Thinking) - Mixed-Form Reasoning: leaderboard

GR-Ben (Thinking) Mixed-Form Reasoning scores LLM critics with slow thinking enabled on the GR-Ben (2026) subset built from the mixed-form logical-reasoning problems (LogiQA 2.0), where the model must name the earliest erroneous step of a step-by-step solution or state that every step is correct. The score is F1 in percent, the harmonic mean of accuracy on flawed and on correct solutions, higher is better.

Metric: F1 (%) of first-error identification: the harmonic mean of the accuracy on flawed solutions (the earliest erroneous step named) and the accuracy on fully correct solutions (judged correct), on the 399 mixed-form logical-reasoning problems (LogiQA 2.0) of GR-Ben (3,595 human-annotated solutions written by 15 LLMs); LLMs prompted as critics with slow thinking enabled (GPT-5.2 at xhigh and Gemini 3 Flash at high reasoning effort, 32,768 output tokens); higher is better. Source: arxiv.org. Saturation forecast: Around March 2028. 7 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1Gemini 3 FlashHigh68#175
2DeepSeek V3.2Thinking54.6#419
3GPT-5.2 (2025-12-11)xHigh52.2
4Qwen 3 8B47.9#1873
5Qwen 3 14B47.3#1269
6Qwen 3 32B46.2#972
7Qwen 3 4B36.5#2348

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

Interactive version: theaggregate.ai/benchmark?slug=gr-ben-thinking-mixed-form-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-21.