L3Cube-IndicQuest v2 (English) - Science: leaderboard

The 325 Science questions of L3Cube-IndicQuest v2 (L3Cube, 2026), India-specific factual questions from curricula and reference books answered closed-book in English and judged by Gemma 3 12B against the gold answer. Accuracy in percent; higher is better.

Metric: Accuracy (%; the 325 Science questions; closed-book short answers to curriculum-grounded India-specific factual questions in English, judged correct or incorrect by Gemma 3 12B against the gold answer (paraphrases and transliterations accepted)). Source: arxiv.org. Saturation forecast: Estimated already saturated. 6 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 2.5 Flash88.6#453
2Gemma 4 31B68.3#394
3GPT-5.4 Mini67.4#389
4Sarvam 30B64.6#1744
5Gemma 2 9B53.2#2802
6Llama 3.1 8B42.2#4861

No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.

Interactive version: theaggregate.ai/benchmark?slug=l3cube-indicquest-v2-english-science · How It Works · Data refreshed daily, snapshot 2026-11-13.