VakQA: leaderboard
Metric: Judge score (1-5) (2,001 Telugu spoken factoid questions from quiz-style interactions across six domains, with human-verified reference answers; each answer scored 1 to 5 for correctness against the reference by Gemini-2.5-Flash as judge (the best-aligned judge in the human study); mean over the questions; oracle input: the human-verified Telugu transcription of each question). Source: arxiv.org. Saturation forecast: Estimated already saturated. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Flash | 3.63 |
| 2 | Gemma 3 27B | 2.55 |
| 3 | Gemma 3 12B | 2.01 |
| 4 | Gemma 3 4B | 1.43 |
| 5 | Qwen 3 4B | 1.17 |
Interactive version: theaggregate.ai/benchmark?slug=vakqa · How It Works · Data refreshed daily, snapshot 2026-09-26.