SEABED - Affective Interpretation (Open-Ended): leaderboard
SEABED (Rajgarhia et al., 2026) is an audio-first reasoning benchmark on Southeast Asian speech built from openly available corpora. In this task the model judges both the valence and the arousal of Thai or Indonesian speech from its prosody; answers are open-ended and a Grok 4.5 judge marks each answer correct or incorrect, and the score is accuracy in percent, higher is better.
Metric: Accuracy (%; 504 joint valence and arousal judgment from prosody items; open-ended answers judged by Grok 4.5, binary correct or incorrect (graded 0 to 1 for long-form); audio at 16 kHz). Source: arxiv.org. Saturation forecast: Around October 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.5 Flash | 37.7 | #98 |
| 2 | Gemini 3.6 Flash | 36.6 | #72 |
| 3 | Gemini 2.5 Pro | 30.8 | #262 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=seabed-affective-interpretation-open-ended · How It Works · Data refreshed daily, snapshot 2026-11-12.