List your company

Audio Question Answering on speech-web-questions (S-Web)

74.9AccuracyQwen3-Omni
43.28451.49259.767.908Apr 8, 2026Apr 11, 2026Apr 15, 2026Apr 19, 2026Apr 22, 2026Apr 26, 2026Apr 30, 2026

Evaluation results

MethodLinks
Qwen3-OmniSize=30B-A3B2026.0474.9
MiniCPM-o 4.5Size=9B2026.0470.2
AudioKVModel=Qwen2.5-Omni-7B, KV cache budget=40%, max_gen_tokens=64, Judge=Gemini 2.5 Flash2026.0455.5
PyramidKVModel=Qwen2.5-Omni-7B, KV cache budget=40%, max_gen_tokens=64, Judge=Gemini 2.5 Flash2026.0455.1
AdaKVModel=Qwen2.5-Omni-7B, KV cache budget=40%, max_gen_tokens=64, Judge=Gemini 2.5 Flash2026.0454.9
SnapKVModel=Qwen2.5-Omni-7B, KV cache budget=40%, max_gen_tokens=64, Judge=Gemini 2.5 Flash2026.0454.5
Kimi-AudioSize=9B2026.0446.4
AudioKVModel=Qwen2.5-Omni-3B, KV cache budget=40%, max_gen_tokens=64, Judge=Gemini 2.5 Flash2026.0446.1
PyramidKVModel=Qwen2.5-Omni-3B, KV cache budget=40%, max_gen_tokens=64, Judge=Gemini 2.5 Flash2026.0445.3
SnapKVModel=Qwen2.5-Omni-3B, KV cache budget=40%, max_gen_tokens=64, Judge=Gemini 2.5 Flash2026.0444.9
AdaKVModel=Qwen2.5-Omni-3B, KV cache budget=40%, max_gen_tokens=64, Judge=Gemini 2.5 Flash2026.0444.5