SPEC CPU result comparison

15th October 2026 - Normalized report

SPEC2026 FPrate: LTO -O2 x86-64-v3 - LLVM and GCC

-O2 -flto -march=x86-64-v3

Please note that the same "optimization levels" mean different things to different compilers. For example, LLVM and LLVM-based compiler tend to be much more aggressive at -O2, often with the corresponding consequences to size of the generated binaries and compilation time. GCC is sometimes more aggressive at -O3. LLVM is also often by default less strict when it comes to standard compliance by default and performs optimizations GCC would only do with a special compilation option or at -Ofast. Keep this is mind when comparing different compilers.

AMD EPYC 8635P (zen5 based Sorano)

Run-times

Benchmark LLVM GCC
709.cactus_r 100.00% 112.75%
722.palm_r 100.00% 148.16%
731.astcenc_r 100.00% 167.84%
736.ocio_r 100.00% 105.19%
737.gmsh_r 100.00% VE
748.flightdm_r 100.00% 100.90%
749.fotonik3d_r 100.00% 101.54%
765.roms_r 100.00% 86.92%
766.femflow_r 100.00% 154.72%
767.nest_r 100.00% 105.00%
772.marian_r 100.00% 878.12%
782.lbm_r 100.00% 69.37%
LLVM clang version 23.1.3 (/labs/mjambor/git/llvm 0d261d1ca552c95a8f007e061c787ac7132fbcbc), flang version 23.1.3 (/labs/mjambor/git/llvm 0d261d1ca552c95a8f007e061c787ac7132fbcbc)
GCC GCC 17.0.0 20261015 (experimental) (revision 84348add236100e5f69a3e9ec4f7b1038bf2b83d)
Naturally, the smaller the run-times, the better.

Non-debug executable sizes

Benchmark LLVM GCC
709.cactus_r 100.00% 87.90%
722.palm_r 100.00% 80.21%
731.astcenc_r 100.00% 99.81%
736.ocio_r 100.00% 87.01%
737.gmsh_r 100.00% VE
748.flightdm_r 100.00% 87.74%
749.fotonik3d_r 100.00% 26.65%
765.roms_r 100.00% 68.73%
766.femflow_r 100.00% 53.36%
767.nest_r 100.00% 82.80%
772.marian_r 100.00% 89.44%
782.lbm_r 100.00% 57.02%
Smaller binary sizes are always better.