ODU-Bench (Audio-Only) - Key-Point Coverage: leaderboard

Metric: Key-point hit rate (%; share of reference key points on structured intent and required context covered by the predicted intent and context, demand scenes only; 731 synthetic audio-only scenes (570 with a demand, 161 without), user turns as audio, earlier assistant replies as text; key points judged by Qwen3.6-Flash with thinking). Source: arxiv.org. Saturation forecast: Around May 2027. 14 models tracked.

Top models

#ModelScore
1Seed 2.0 Lite78
2Qwen 3.5 Omni Plus71.5
3Gemini 3.1 Pro (Preview)70.1
4Gemini 3.7 Flash66.1
5Qwen3 Omni 30B A3B Instruct60.9
6Gemini 3.5 Flash Lite56.4
7MiniCPM-o-4.547.9
8Qwen2.5-Omni-7B45.7

Interactive version: theaggregate.ai/benchmark?slug=odu-bench-audio-only-key-point-coverage · How It Works · Data refreshed daily, snapshot 2026-09-26.