SPEC CPU result comparison

15th October 2026 - Normalized report

SPEC2026 FPrate: PGO generic -O2 - Released GCCs vs Trunk

-O2 -fprofile-use

AMD EPYC 8635P (zen5 based Sorano)

Run-times

Benchmark GCC 16 Trunk
709.cactus_r 100.00% 105.99%
722.palm_r 100.00% 97.33%
731.astcenc_r 100.00% 98.52%
736.ocio_r 100.00% 99.35%
737.gmsh_r 100.00% 98.29%
748.flightdm_r 100.00% 99.08%
749.fotonik3d_r 100.00% 102.71%
765.roms_r 100.00% 100.00%
766.femflow_r 100.00% 93.12%
767.nest_r 100.00% 101.48%
772.marian_r 100.00% 100.00%
782.lbm_r 100.00% 99.56%
GCC 15 GCC 15.3.0 (revision 4db0e8df15bef836558857c291c323add11d035c)
GCC 16 GCC 16.2.0 (revision 78d4ac73dd391005b895a6148cd9831e28e1208b)
Trunk GCC 17.0.0 20261015 (experimental) (revision 84348add236100e5f69a3e9ec4f7b1038bf2b83d)
Naturally, the smaller the run-times, the better.
See also continuous gcc branch benchmarking at https://lnt.opensuse.org/db_default/v4/SPEC/spec_report/branch.

Non-debug executable sizes

Benchmark GCC 16 Trunk
709.cactus_r 100.00% 100.65%
722.palm_r 100.00% 99.71%
731.astcenc_r 100.00% 100.27%
736.ocio_r 100.00% 100.79%
737.gmsh_r 100.00% 99.81%
748.flightdm_r 100.00% 100.14%
749.fotonik3d_r 100.00% 99.06%
765.roms_r 100.00% 99.23%
766.femflow_r 100.00% 100.41%
767.nest_r 100.00% 100.56%
772.marian_r 100.00% 101.20%
782.lbm_r 100.00% 99.48%
Smaller binary sizes are always better.