SEABED - Affective Interpretation (Open-Ended): leaderboard

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 August 2028. 6 models tracked.

Top models

#ModelScore
1Gemini 3.5 Flash37.7
2Gemini 3.6 Flash36.6
3Gemini 2.5 Pro30.8
4SeaLLMs-Audio-7B25

Interactive version: theaggregate.ai/benchmark?slug=seabed-affective-interpretation-open-ended · How It Works · Data refreshed daily, snapshot 2026-09-26.