MuLA-Bench - Italian: leaderboard
Metric: Accuracy (%; the 304 Italian-language questions over both evidence tracks; open-ended questions over in-the-wild long-form recordings (median 36.8 minutes, capped at three hours) in the source language, answered from the full recording without transcripts, audio clipped only to a local model's native input limit; each answer judged correct or not by GPT-5.6 Sol on a route-specific binary rubric with programmatic structural and 3-second timing checks; missing or invalid answers scored wrong). Source: arxiv.org. Saturation forecast: Around December 2026. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.8 Flash | 71.1 |
| 2 | Gemini 3.7 Flash | 65.1 |
| 3 | Qwen 3.5 Omni Plus | 56.2 |
| 4 | Gemini 3.1 Pro (Preview) | 55.6 |
| 5 | Qwen3.5 Omni Flash | 38.2 |
| 6 | Muse Spark 1.2 | 33.2 |
Interactive version: theaggregate.ai/benchmark?slug=mula-bench-italian · How It Works · Data refreshed daily, snapshot 2026-09-26.