GSM-Plus-BN - Critical Thinking (CoT): leaderboard

Metric: Accuracy (%; critical thinking perturbation, essential information removed, so the model must say the question cannot be answered; Bangla grade-school math word problems translated from GSM-Plus and verified by human translators; chain-of-thought prompting; Groq API). Source: arxiv.org. Saturation forecast: Around January 2027. 6 models tracked.

Top models

#ModelScore
1GPT-OSS-20B33
2GPT-OSS-120B31.8
3Llama 4 Scout Instruct27.4
4Qwen 3 32B11

Interactive version: theaggregate.ai/benchmark?slug=gsm-plus-bn-critical-thinking-cot · How It Works · Data refreshed daily, snapshot 2026-09-29.