GSM-Plus-BN - Digit Expansion: leaderboard
GSM-Plus-BN (2026) is a human-verified Bangla translation of the GSM-Plus perturbation benchmark for grade-school math. Accuracy in percent on the digit expansion variants (numbers expanded to more digits) with standard prompting; higher is better.
Metric: Accuracy (%; digit expansion perturbation, numbers expanded to more digits; Bangla grade-school math word problems translated from GSM-Plus and verified by human translators; standard prompting (answer directly); Groq API). Source: arxiv.org. Saturation forecast: Estimated already saturated. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-OSS-20B | 82.6 | #1147 |
| 2 | GPT-OSS-120B | 82.4 | #700 |
| 3 | Llama 4 Scout Instruct | 81.6 | #1543 |
| 4 | Qwen 3 32B | 39.2 | #960 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=gsm-plus-bn-digit-expansion · How It Works · Data refreshed daily, snapshot 2026-11-13.