ResearchBenchmarksLatency Profiling on Qwen 7B Nano 8G 2.5 (inference)Follow57.7TPOT (ms)UniQL54.66475.15795.65116.143Dec 3, 2025Evaluation ResultsMethodMethodLinksTPOT (ms)TTLT (ms)UniQLW-bit=4, Prun. p%=35%W-bit=4, Prun. p%=35%2025.1257.728,185.8UniQLW-bit=4, Prun. p%=0%W-bit=4, Prun. p%=0%2025.1277.239,795.3TAO-HQQW-bit=4*, Prun. p%=0%W-bit=4*, Prun. p%=0%2025.12133.680,770.2