AudioBench - IMDA 30s Dialogue Summarization Human — leaderboard
Metric: Mean Score (%). Source: github.com. 5 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct | 50.15 |
| 2 | cascade_whisper_large_v3_llama_3_8b_instruct | 49.25 |
| 3 | Qwen2-Audio-7B-Instruct | 37.6 |
| 4 | Qwen-Audio-Chat | 30.65 |
| 5 | SALMONN_7B | 16.15 |
Interactive version: theaggregate.ai/benchmark?slug=audiobench-imda-30s-dialogue-summarization-human · How the rankings work · Data refreshed daily, snapshot 2026-07-22.