SPEC CPU result comparison

15th October 2026 - Normalized report

SPEC2017 INTrate: PGO+LTO -O2 generic - LLVM and GCC

-O2 -flto and -fprofile-instr-use vs. -profile-use

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
500.perlbench_r 100.00% 84.94%
502.gcc_r 100.00% 101.84%
505.mcf_r 100.00% 83.04%
520.omnetpp_r 100.00% 81.78%
523.xalancbmk_r 100.00% 87.47%
525.x264_r 100.00% 92.70%
531.deepsjeng_r 100.00% 105.41%
541.leela_r 100.00% 101.91%
548.exchange2_r CE
557.xz_r 100.00% 96.23%
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
500.perlbench_r 100.00% 95.66%
502.gcc_r 100.00% 80.98%
505.mcf_r 100.00% 89.00%
520.omnetpp_r 100.00% 79.95%
523.xalancbmk_r 100.00% 81.35%
525.x264_r 100.00% 94.14%
531.deepsjeng_r 100.00% 102.43%
541.leela_r 100.00% 93.39%
548.exchange2_r CE
557.xz_r 100.00% 95.68%
Smaller binary sizes are always better.

AMD EPYC 75F3 (zen3 based high-frequency Milan)

Run-times

Benchmark LLVM GCC
500.perlbench_r 100.00% 80.51%
502.gcc_r 100.00% 101.19%
505.mcf_r 100.00% 84.05%
520.omnetpp_r 100.00% 85.48%
523.xalancbmk_r 100.00% 93.09%
525.x264_r 100.00% 87.76%
531.deepsjeng_r 100.00% 105.85%
541.leela_r 100.00% 104.10%
548.exchange2_r CE
557.xz_r 100.00% 93.80%
LLVM clang version 23.1.3 (https://github.com/llvm/llvm-project.git 0d261d1ca552c95a8f007e061c787ac7132fbcbc), flang version 23.1.3 (https://github.com/llvm/llvm-project.git 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
500.perlbench_r 100.00% 95.66%
502.gcc_r 100.00% 80.99%
505.mcf_r 100.00% 89.39%
520.omnetpp_r 100.00% 79.92%
523.xalancbmk_r 100.00% 81.35%
525.x264_r 100.00% 94.14%
531.deepsjeng_r 100.00% 102.43%
541.leela_r 100.00% 93.45%
548.exchange2_r CE
557.xz_r 100.00% 95.69%
Smaller binary sizes are always better.