IndicIFEval-Trans - Bengali: leaderboard
Metric: Prompt-level loose accuracy (%, 0-100) on the Bengali prompts; IndicIFEval-Trans: IFEval prompts translated with Sarvam-Translate, localized and verified by native speakers; the 321 prompts valid in every language; IFEval-style verifiable instructions checked by the IFEval rules with Indic NLP Library tokenization and sentence splitting; prompt-level loose accuracy (a prompt counts when all of its instructions are followed, after minor formatting relaxations); greedy decoding through the LM Evaluation Harness; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 22 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro | 92.5 | #77 |
| 2 | GPT-5 | 90.7 | #91 |
| 3 | Gemini 3 Flash | 90 | #93 |
| 4 | GPT-5 Mini (2025-08-07) | 86 | #165 |
| 5 | Gemma 3 27B (IT) | 81 | #509 |
| 6 | Gemma 3 12B (IT) | 78.8 | #655 |
| 7 | Llama 3.3 70B Instruct | 77 | #520 |
| 8 | Llama 4 Scout Instruct | 76 | #526 |
| 9 | Gemma 3 4B (IT) | 73.8 | #971 |
| 10 | Qwen 3 32B | 73.8 | #424 |
| 11 | Qwen 3 14B | 72.6 | #524 |
| 12 | Qwen 3 8B | 67.3 | #667 |
| 13 | Llama 3.1 70B Instruct | 67 | #548 |
| 14 | Qwen 3 4B | 62.6 | #823 |
| 15 | Llama 3.1 8B Instruct | 44.9 | #1018 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=indicifeval-trans-bengali · How It Works · Data refreshed daily, snapshot 2026-10-11.