List your company
ResearchFormerly WizWand

Audio Question Answering on MMAU

77.5ScoreQwen3-Omni (Sep 25)
56.59662.02367.4572.877Oct 14, 2025Nov 15, 2025Dec 18, 2025Jan 19, 2026Feb 21, 2026Mar 25, 2026Apr 27, 2026
Updated 6mo ago

Evaluation results

MethodLinks
Qwen3-Omni (Sep 25)Model Category=Omnimodal MLLMs2026.0477.5
Gemini 2.5 Pro (May 25)Model Category=Closed-Source APIs2026.0477.4
Omni-o3Model Category=Omnimodal MLLMs2026.0475.6
AFlamingo3 (NeurIPS 25)Model Category=Audio MLLMs2026.0472.4
Gemini 2.5 ProModel Category=Proprietary Models, QA Backbone=GPT-4o, Evaluation Protocol=caption-to-QA cascade2025.1070
Audio-Captioner-7BModel Category=Open-Source Models, QA Backbone=GPT-4o, Evaluation Protocol=caption-to-QA cascade2025.1070
Gemini 2.5 ProModel Category=Proprietary Models, QA Backbone=GPT-4o, Evaluation Protocol=caption-to-QA cascade2025.1070
Audio-Captioner-7BModel Category=Open-Source Models, QA Backbone=GPT-4o, Evaluation Protocol=caption-to-QA cascade2025.1070
Ola (Feb 25)Model Category=Omnimodal MLLMs2026.0467.1
Gemini 2.5 FlashModel Category=Proprietary Models, QA Backbone=GPT-4o, Evaluation Protocol=caption-to-QA cascade2025.1065.6
Gemini 2.5 FlashModel Category=Proprietary Models, QA Backbone=GPT-4o, Evaluation Protocol=caption-to-QA cascade2025.1065.6
Qwen2.5-Omni (Mar 25)Model Category=Omnimodal MLLMs2026.0465.5
Qwen2.5-Omni-7BModel Category=Open-Source Models, QA Backbone=GPT-4o, Evaluation Protocol=caption-to-QA cascade2025.1065.2
Qwen2.5-Omni-7BModel Category=Open-Source Models, QA Backbone=GPT-4o, Evaluation Protocol=caption-to-QA cascade2025.1065.2
DeSTA 2.5 (Jul 25)Model Category=Audio MLLMs2026.0465.2
AudioReaonser (Sep 25)Model Category=Audio MLLMs2026.0463.8
Qwen2-Audio-7BModel Category=Open-Source Models, QA Backbone=GPT-4o, Evaluation Protocol=caption-to-QA cascade2025.1063.3
Qwen2-Audio-7BModel Category=Open-Source Models, QA Backbone=GPT-4o, Evaluation Protocol=caption-to-QA cascade2025.1063.3
GPT-4o (May 24)Model Category=Closed-Source APIs2026.0462.5
GPT-4o AudioModel Category=Proprietary Models, QA Backbone=GPT-4o, Evaluation Protocol=caption-to-QA cascade2025.1062.4
GPT-4o AudioModel Category=Proprietary Models, QA Backbone=GPT-4o, Evaluation Protocol=caption-to-QA cascade2025.1062.4
MiDashengLM-7BModel Category=Open-Source Models, QA Backbone=GPT-4o, Evaluation Protocol=caption-to-QA cascade2025.1059.4
MiDashengLM-7BModel Category=Open-Source Models, QA Backbone=GPT-4o, Evaluation Protocol=caption-to-QA cascade2025.1059.4
Gemini 2.0 FlashModel Category=Proprietary Models, QA Backbone=GPT-4o, Evaluation Protocol=caption-to-QA cascade2025.1058.6
Gemini 2.0 FlashModel Category=Proprietary Models, QA Backbone=GPT-4o, Evaluation Protocol=caption-to-QA cascade2025.1058.6
SALMONN-13BModel Category=Open-Source Models, QA Backbone=GPT-4o, Evaluation Protocol=caption-to-QA cascade2025.1058.36
SALMONN-13BModel Category=Open-Source Models, QA Backbone=GPT-4o, Evaluation Protocol=caption-to-QA cascade2025.1058.36
Qwen2-Audio (Jul 24)Model Category=Audio MLLMs2026.0457.4