CASU - Counterfactual Reasoning: leaderboard
Metric: Accuracy (%; four-option multiple-choice questions on semi-synthetic auditory scenes that combine speech, sound events and background; counterfactual reasoning: how an interpretation changes if the acoustic evidence were altered). Source: arxiv.org. Saturation forecast: Around December 2026. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o Audio | 74.96 |
| 2 | Qwen3 Omni 30B A3B Instruct | 74.5 |
| 3 | Gemini 2.0 Flash | 73.83 |
| 4 | Qwen2.5-Omni-7B | 65.38 |
| 5 | Voxtral-Small-24B-2507 | 55.99 |
| 6 | SALMONN-13B | 46.95 |
| 7 | Qwen2-Audio-7B-Instruct | 37.81 |
Interactive version: theaggregate.ai/benchmark?slug=casu-counterfactual-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-26.