SPEC CPU result comparison

15th October 2026 - Normalized report

SPEC2026 INTrate: PGO -O2 generic - LLVM and GCC

-O2 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
706.stockfish_r 100.00% 43.89%
707.ntest_r 100.00% 107.83%
708.sqlite_r 100.00% 96.97%
710.omnetpp_r 100.00% 92.06%
714.cpython_r 100.00% 94.72%
721.gcc_r 100.00% 99.35%
723.llvm_r 100.00% 106.16%
727.cppcheck_r 100.00% 104.74%
729.abc_r 100.00% 100.85%
734.vpr_r 100.00% 98.92%
735.gem5_r 100.00% 98.19%
750.sealcrypto_r 100.00% 111.56%
753.ns3_r 100.00% 101.86%
777.zstd_r 100.00% 92.31%
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
706.stockfish_r 100.00% 102.18%
707.ntest_r 100.00% 96.54%
708.sqlite_r 100.00% 97.27%
710.omnetpp_r 100.00% 78.57%
714.cpython_r 100.00% 85.86%
721.gcc_r 100.00% 91.18%
723.llvm_r 100.00% 88.92%
727.cppcheck_r 100.00% 90.52%
729.abc_r 100.00% 90.37%
734.vpr_r 100.00% 87.84%
735.gem5_r 100.00% 85.91%
750.sealcrypto_r 100.00% 89.71%
753.ns3_r 100.00% 78.92%
777.zstd_r 100.00% 89.42%
Smaller binary sizes are always better.