L3Cube-IndicQuest v2 (English): leaderboard

Metric: Weighted accuracy (%; all 3,471 English questions, domains weighted by size; 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: Around December 2026. 6 models tracked.

Top models

#ModelScore
1Gemini 2.5 Flash86.9
2Gemma 4 31B62.2
3GPT-5.4 Mini60.1
4Sarvam 30B48.7
5Gemma 2 9B41.1
6Llama 3.1 8B35.4

Interactive version: theaggregate.ai/benchmark?slug=l3cube-indicquest-v2-english · How It Works · Data refreshed daily, snapshot 2026-09-29.