ResearchBenchmarksModel Compression on Qwen 7B 2.5Model Compression on Qwen 7B 2.5Follow2.7Model Size (GB)UniQL2.25.5758.9512.325Dec 3, 2025Evaluation resultsMethodMethodLinksModel Size (GB)UniQLW-bit=4, Prun. p%=35%W-bit=4, Prun. p%=35%2025.122.7UniQLW-bit=4, Prun. p%=0%W-bit=4, Prun. p%=0%2025.123.9TRT-AWQW-bit=4*, Prun. p%=0%W-bit=4*, Prun. p%=0%2025.125.6TAO-HQQW-bit=4*, Prun. p%=0%W-bit=4*, Prun. p%=0%2025.126MoDeGPTW-bit=16, Prun. p%=15%W-bit=16, Prun. p%=15%2025.1213.2SVD-LLMW-bit=16, Prun. p%=15%W-bit=16, Prun. p%=15%2025.1213.3FP16W-bit=16, Prun. p%=0%W-bit=16, Prun. p%=0%2025.1215.2