ESCUCHA: leaderboard

Metric: Accuracy (%; all 1,000 questions: 900 multiple-choice items scored by exact letter match and 100 audio instruction-following items checked by a deterministic verifier; in-the-wild Spanish audio from 6.7 s to 85 min per question with multiple accents and non-normative speech; one shared system prompt for every model; random baseline 23.05%). Source: arxiv.org. Saturation forecast: Around 2031. 8 models tracked.

Top models

#ModelScore
1Gemini 2.5 Flash52.8
2Qwen2.5-Omni-7B50.6
3Gemma 4 12B (IT)49.3
4Voxtral-Mini-3B-250749.2

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