SPEC CPU result comparison

15th October 2026 - Normalized report

SPEC2026 FPrate: Standard -O2 generic - LLVM and GCC

Plain -O2

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% 105.25%
722.palm_r 100.00% 122.07%
731.astcenc_r 100.00% 146.34%
736.ocio_r 100.00% 109.07%
737.gmsh_r 100.00% 94.50%
748.flightdm_r 100.00% 94.58%
749.fotonik3d_r 100.00% 95.35%
765.roms_r 100.00% 89.80%
766.femflow_r 100.00% 484.04%
767.nest_r 100.00% 99.14%
772.marian_r 100.00% 401.39%
782.lbm_r 100.00% 96.15%
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% 103.31%
722.palm_r 100.00% 83.49%
731.astcenc_r 100.00% 121.05%
736.ocio_r 100.00% 114.82%
737.gmsh_r 100.00% 101.37%
748.flightdm_r 100.00% 106.92%
749.fotonik3d_r 100.00% 28.14%
765.roms_r 100.00% 71.70%
766.femflow_r 100.00% 75.32%
767.nest_r 100.00% 103.48%
772.marian_r 100.00% 105.69%
782.lbm_r 100.00% 86.33%
Smaller binary sizes are always better.