List your company
ResearchFormerly WizWand

Audio Question Answering on speech-chatbot-alpaca-eval (S-Alpaca)

66.7AccuracyAudioKV
54.01257.30660.663.894Apr 8, 2026
Updated 7mo ago

Evaluation results

MethodLinks
AudioKVModel=Qwen2.5-Omni-7B, KV cache budget=40%, max_gen_tokens=64, Judge=Gemini 2.5 Flash2026.0466.7
PyramidKVModel=Qwen2.5-Omni-7B, KV cache budget=40%, max_gen_tokens=64, Judge=Gemini 2.5 Flash2026.0465.2
SnapKVModel=Qwen2.5-Omni-7B, KV cache budget=40%, max_gen_tokens=64, Judge=Gemini 2.5 Flash2026.0464.7
AdaKVModel=Qwen2.5-Omni-7B, KV cache budget=40%, max_gen_tokens=64, Judge=Gemini 2.5 Flash2026.0464.7
AudioKVModel=Qwen2.5-Omni-3B, KV cache budget=40%, max_gen_tokens=64, Judge=Gemini 2.5 Flash2026.0457.6
AdaKVModel=Qwen2.5-Omni-3B, KV cache budget=40%, max_gen_tokens=64, Judge=Gemini 2.5 Flash2026.0456.1
PyramidKVModel=Qwen2.5-Omni-3B, KV cache budget=40%, max_gen_tokens=64, Judge=Gemini 2.5 Flash2026.0456.1
SnapKVModel=Qwen2.5-Omni-3B, KV cache budget=40%, max_gen_tokens=64, Judge=Gemini 2.5 Flash2026.0454.5