ResearchBenchmarksLatency Profiling on Nemotron-H-8BLatency Profiling on Nemotron-H-8BFollow6.8TPOTUniQL6.09610.84815.620.352Dec 3, 2025Evaluation resultsMethodMethodLinksTPOTTTLTUniQLW-bit=4, Prun. p%=35%W-bit=4, Prun. p%=35%2025.126.86,955.6UniQLW-bit=4, Prun. p%=0%W-bit=4, Prun. p%=0%2025.128.29,095.7FP16W-bit=16, Prun. p%=0%W-bit=16, Prun. p%=0%2025.1224.425,889.4