SPEC CPU®2026 Floating Point Rate Result

Copyright 2026 Standard Performance Evaluation Corporation

Apple (Test Sponsor: SPEC CPU Supporting Contributor jhenning)

MacBook Pro (16-inch, M5 Pro) 1-copy SPECrate

SPECrate®2026_fp_base = 10.90

SPECrate®2026_fp_peak = Not Run

CPU2026 License: 7 Test Date: Apr-2026
Test Sponsor: SPEC CPU Supporting Contributor jhenning Hardware Availability: Mar-2026
Tested by: SPEC CPU Supporting Contributor jhenning Software Availability: Apr-2026
Benchmark results graph
Hardware
CPU Name: Apple M5 Pro
  Max MHz: 4600 (approximate; see notes)
  Nominal: 4600 (approximate; see notes)
Enabled: 18 (6 Super and 12 Performance) cores, 1 chip
Orderable: 1 chip
Cache L1: 192 KB I + 128 KB D on chip per core (Super);
128 KB I + 64 KB D on chip per core (Performance)
  L2: 32 MB I+D on chip per chip,
(16 MB per 6 Super cores, 8 MB per 6
Performance cores)
  L3: None disclosed
  Other: None disclosed
Memory: 64.0 GB
Storage: 4 TB APPLE SSD AP4096Z
Cooling: Air
Other: None
Software
OS: macOS 26.4.1 (25E253)
Compiler: C/C++: Version 22.1.3 of LLVM;
Fortran: Version 22.1.3 of Homebrew flang
Compiler Category: Community
Firmware: Version 18000.101.7 released Apr-2026
File System: apfs
System State: Default
Base Pointers: 64-bit
Peak Pointers: Not Applicable
Other: Xcode 26.4.1.0.1775747724
Power Management: Default

Results Table

Benchmark Base Peak
Copies Seconds Ratio Seconds Ratio Seconds Ratio Copies Seconds Ratio Seconds Ratio Seconds Ratio
709.cactus_r 1 1100 7.77 1130 7.60 1130 7.57
722.palm_r 1 96.6 13.70 97.3 13.60 97.3 13.60
731.astcenc_r 1 75.1 11.20 76.2 11.00 76.4 11.00
736.ocio_r 1 69.9 12.50 70.6 12.40 70.8 12.40
737.gmsh_r 1 75.5 6.08 75.4 6.09 75.7 6.06
748.flightdm_r 1 75.3 9.51 76.4 9.37 76.1 9.41
749.fotonik3d_r 1 72.8 15.90 72.9 15.80 73.0 15.80
765.roms_r 1 1270 12.40 1280 12.30 1280 12.30
766.femflow_r 1 1010 14.60 1010 14.50 99.9 14.70
767.nest_r 1 74.8 10.60 75.2 10.50 74.4 10.70
772.marian_r 1 1080 14.60 1090 14.50 1080 14.60
782.lbm_r 1 76.0 7.54 76.0 7.54 75.9 7.55
SPECrate®2026_fp_base 10.90
SPECrate®2026_fp_peak Not Run
Results appear in the order in which they were run. Bold underlined text indicates a median measurement.

Compiler Notes


 NOTE - MODERATE COMPILER TUNING

 The results on this page intentionally use only moderate compiler tuning. This page differs from many other SPEC CPU
 results which are published to demonstrate the best performance that can be achieved by a system. Such tests often
 use compilers that are optimized for specific chip vendors and use advanced tuning options.

 Moderate optimization may be of interest to users who prefer a simple build process, or who do not know whether
 their application works correctly with advanced tuning options. Specifically, these tests:

    - Use an open source, community-supported compiler, rather than a vendor-specific compiler.
    - Specify only one optimization option, namely -O3.
    - Avoid options that disregard strict standards compliance, such as -Ofast, -ffinite-math-only, or
      -funsafe-math-optimizations.
    - Avoid options such as -march=native or -mcpu=native. Without those, the code generator cannot assume that
      the program will run only on systems that use the same chip as the current system.

 There are many other kinds of testing can be done with SPEC CPU, such as:
   - Performance modeling prior to CPU chip development.
   - Hardware validation.
   - Compiler regression testing (both correctness and performance).
   - Academic and industrial research into new optimizations for CPUs, memory systems, and compilers.

General Notes


 Yes: The test sponsor attests, as of date of publication, that CVE-2017-5715
      (Spectre variant 2) is mitigated in the system as tested and documented.
 Yes: The test sponsor attests, as of date of publication, that CVE-2017-5753
      (Spectre variant 1) is mitigated in the system as tested and documented.
 Yes: The test sponsor attests, as of date of publication, that CVE-2017-5754
      (Meltdown) is mitigated in the system as tested and documented.
 The above 3 statements are based on https://support.apple.com/en-us/101886

Platform Notes


 Sysinfo program /Volumes/ScratchVolume/spec/cpu2026/1.0.0/bin/sysinfo
 Rev: 779ab21020787073335a329f3a45e2cd
 running on i4gots-MacBook-Pro Wed Apr 22 02:07:57 2026

 SUT (System Under Test) info as seen by some common utilities.

 uname -srvm
   Darwin 25.4.0 Darwin Kernel Version 25.4.0: Thu Mar 19 19:33:50 PDT 2026;
   root:xnu-12377.101.15~1/RELEASE_ARM64_T6050 arm64

 w
    2:07  up  6:58, 3 users, load averages: 11.95 19.62 21.28
   USER       TTY      FROM    LOGIN@  IDLE WHAT
   i4got      console  -      Tue19    6:57 -
   i4got      s000     -      Tue19    2:14 -bash
   i4got      s001     -      Tue19    2:13 tail -f reportable.fprate.18-copies.try-1.out

 Username
   From environment variable $USER:  i4got

 ulimit -a
   core file size          (blocks, -c) 0
   data seg size           (kbytes, -d) unlimited
   file size               (blocks, -f) unlimited
   max locked memory       (kbytes, -l) unlimited
   max memory size         (kbytes, -m) unlimited
   open files                      (-n) 256
   pipe size            (512 bytes, -p) 1
   stack size              (kbytes, -s) 65000
   cpu time               (seconds, -t) unlimited
   max user processes              (-u) 10666
   virtual memory          (kbytes, -v) unlimited

 sysinfo process ancestry
  /sbin/launchd
  /System/Applications/Utilities/Terminal.app/Contents/MacOS/Terminal
  login -pf i4got
  -bash
  /bin/bash ./run.5.sh
  runcpu -c clang5 -n 3 --reportable --nobuild --copies=1 fprate
  specperl $SPEC/bin/sysinfo
 $SPEC = /Volumes/ScratchVolume/spec/cpu2026/1.0.0

 Hardware:

     Hardware Overview:

       Model Name: MacBook Pro
       Model Identifier: Mac17,8
       Model Number: Z1N0001A3LL/A
       Chip: Apple M5 Pro
       Total Number of Cores: 18 (6 Super and 12 Performance)
       Memory: 64 GB
       System Firmware Version: 18000.101.7
       OS Loader Version: 18000.101.7

 Software:

     System Software Overview:

       System Version: macOS 26.4.1 (25E253)
       Kernel Version: Darwin 25.4.0
       Boot Volume: Macintosh HD
       Boot Mode: Normal
       Computer Name: i4got's MacBook Pro
       User Name: i4got (i4got)
       Secure Virtual Memory: Enabled
       System Integrity Protection: Enabled
       Time since boot: 6 hours, 58 minutes

 Memory:

       Memory: 64 GB
       Type: LPDDR5
       Manufacturer: Micron

 Developer:

     Developer Tools:

       Version: 26.4.1 (17E202)
       Location: /Applications/Xcode.app
       Applications:
           Xcode: 26.4.1 (24909.0.3)
           Instruments: 26.4.1 (64575.156)
       SDKs:
           DriverKit:
               25.4:
           iOS:
               26.4: (23E252)
           iOS Simulator:
               26.4: (23E252)
           macOS:
               26.4: (25E251)
           tvOS:
               26.4: (23L236)
           tvOS Simulator:
               26.4: (23L236)
           visionOS:
               26.4: (23O248)
           visionOS Simulator:
               26.4: (23O248)
           watchOS:
               26.4: (23T238)
           watchOS Simulator:
               26.4: (23T238)


 From sysctl:
    machdep.cpu.core_count: 18
    machdep.cpu.thread_count: 18
    machdep.cpu.brand_string: Apple M5 Pro
    hw.memsize: 68719476736
    hw.activecpu: 18
    hw.perflevel0.l1icachesize: 196608
    hw.perflevel0.l1dcachesize: 131072
    hw.perflevel0.l2cachesize: 16777216
    hw.perflevel1.l1icachesize: 131072
    hw.perflevel1.l1dcachesize: 65536
    hw.perflevel1.l2cachesize: 8388608
    hw.physicalcpu: 18
    hw.physicalcpu_max: 18
    hw.logicalcpu: 18
    hw.logicalcpu_max: 18
    hw.cacheconfig: 18 1 6 0 0 0 0 0 0 0
    hw.cachesize: 2995912704 65536 8388608 0 0 0 0 0 0 0
    hw.pagesize: 16384
    hw.pagesize32: 16384
    hw.cachelinesize: 128
    hw.l1icachesize: 131072
    hw.l1dcachesize: 65536
    hw.l2cachesize: 8388608
    hw.memsize_usable: 67420422144

 SPEC is set to: /Volumes/ScratchVolume/spec/cpu2026/1.0.0
    Filesystem      Size    Used   Avail Capacity iused ifree %iused  Mounted on
    /dev/disk3s1   1998G    570G   1428G    29%    1.3M   14G    0%   /Volumes/ScratchVolume

 (End of data from sysinfo program)


 ADDITIONAL INFORMATION FROM SYSCTL

  $ sysctl hw | grep  perflevel0
  hw.perflevel0.physicalcpu: 6
  hw.perflevel0.physicalcpu_max: 6
  hw.perflevel0.logicalcpu: 6
  hw.perflevel0.logicalcpu_max: 6
  hw.perflevel0.l1icachesize: 196608
  hw.perflevel0.l1dcachesize: 131072
  hw.perflevel0.l2cachesize: 16777216
  hw.perflevel0.cpusperl2: 6
  hw.perflevel0.name: Super
  $ sysctl hw | grep  perflevel1
  hw.perflevel1.physicalcpu: 12
  hw.perflevel1.physicalcpu_max: 12
  hw.perflevel1.logicalcpu: 12
  hw.perflevel1.logicalcpu_max: 12
  hw.perflevel1.l1icachesize: 131072
  hw.perflevel1.l1dcachesize: 65536
  hw.perflevel1.l2cachesize: 8388608
  hw.perflevel1.cpusperl2: 6
  hw.perflevel1.name: Performance
  $

 The above shows cache sizes for both core types, and "cpusperl2" indicates sharing among cores. See
 https://developer.apple.com/documentation/kernel/1387446-sysctlbyname/determining_system_capabilities
 Note: it is possible that the sysctl cache information may be incomplete.  There might be other cache
 types or other cache characteristics that are not disclosed by the above 'sysctl' command.


 ADDITIONAL INFORMATION FROM POWERMETRICS

 Regarding the CPU MHz fields above, note that:
  - The vendor does not state CPU frequencies in their marketing materials.
  - Frequency levels are printed by the vendor-provided 'powermetrics' utility; these might or might
    not correspond to traditional "nominal" or "maximum" MHz.  Therefore, the values on this page for
    Nominal and Maximum MHz must be considered approximate.
  - The table below lists all of the MHz values mentioned in the dvfm_states section of powermetrics.
  - On the tested system, the maximum frequencies mentioned are 4608 for the S-cluster and 4380 for the
    P-Clusters

 ----------------------------------------------------------------------------------------------------------
 CPU Dynamic Frequencies                                                 From 'powermetrics --format=plist'
 ----------------------------------------------------------------------------------------------------------
 processor
 clusters    dvfm_states
 name        freq
 ----------  ----------------------------------------------------------------------------------------------
 P0-Cluster  1344 1644 1992 2304 2652 2964 3240 3504 3696 3876 4044 4176 4284 4308 4380
 P1-Cluster  1344 1644 1992 2304 2652 2964 3240 3504 3696 3876 4044 4176 4284 4308 4380
 S-Cluster   1308 1620 1980 2292 2580 2880 3180 3432 3648 3828 3984 4104 4188 4236 4284 4308 4332 4428 4512 4608

 The table above was extracted from powermetrics via Perl code:

 open P, "powermetrics --sample-count=0  --show-initial-usage --samplers=cpu_power --format=plist|"
    or die "cannot run powermetrics";
 while (<P>) {
    if (m{<key>name</key><string>(\S+Cluster)</string>}) {
       printf "%-10s  ", $1;
       $printing = 1;
       @freqs    = ();
    }
    next if ! $printing;
    if (m{^</array>}) {
       print join(" ", @freqs), "\n";
       $printing = 0;
    }
    if (m{<key>freq</key><integer>(\d+)</integer>}) {
       push @freqs, $1;
    }
 }

Compiler Version Notes

==============================================================================
C       | 782.lbm_r(base)
------------------------------------------------------------------------------
Homebrew clang version 22.1.3
Target: arm64-apple-darwin25.4.0
Thread model: posix
InstalledDir: /opt/homebrew/Cellar/llvm/22.1.3/bin
Configuration file:
  /opt/homebrew/Cellar/llvm/22.1.3/etc/clang/arm64-apple-darwin25.cfg
------------------------------------------------------------------------------

==============================================================================
C++     | 731.astcenc_r(base) 736.ocio_r(base) 748.flightdm_r(base)
        | 766.femflow_r(base) 767.nest_r(base) 772.marian_r(base)
------------------------------------------------------------------------------
Homebrew clang version 22.1.3
Target: arm64-apple-darwin25.4.0
Thread model: posix
InstalledDir: /opt/homebrew/Cellar/llvm/22.1.3/bin
Configuration file:
  /opt/homebrew/Cellar/llvm/22.1.3/etc/clang/arm64-apple-darwin25.cfg
------------------------------------------------------------------------------

==============================================================================
C++, C  | 709.cactus_r(base) 737.gmsh_r(base)
------------------------------------------------------------------------------
Homebrew clang version 22.1.3
Target: arm64-apple-darwin25.4.0
Thread model: posix
InstalledDir: /opt/homebrew/Cellar/llvm/22.1.3/bin
Configuration file:
  /opt/homebrew/Cellar/llvm/22.1.3/etc/clang/arm64-apple-darwin25.cfg
------------------------------------------------------------------------------

==============================================================================
Fortran | 722.palm_r(base) 749.fotonik3d_r(base) 765.roms_r(base)
------------------------------------------------------------------------------
Homebrew flang version 22.1.3
Target: arm64-apple-darwin25.4.0
Thread model: posix
InstalledDir: /opt/homebrew/Cellar/flang/22.1.3/libexec
Configuration file: /opt/homebrew/Cellar/flang/22.1.3/libexec/flang.cfg
Configuration file:
  /opt/homebrew/Cellar/flang/22.1.3/etc/clang/arm64-apple-darwin25.cfg
------------------------------------------------------------------------------

Base Compiler Invocation

C benchmarks:

 clang 

C++ benchmarks:

 clang++ 

Fortran benchmarks:

 flang 

Benchmarks using both C and C++:

 clang++   clang 

Base Portability Flags

766.femflow_r:  -DSPEC_NO_CXX17_SPECIAL_MATH_FUNCTIONS 

Base Optimization Flags

C benchmarks:

 -std=c18   -g   -O3   -Wno-implicit-function-declaration 

C++ benchmarks:

 -std=c++17   -g   -O3   -Wno-implicit-function-declaration 

Fortran benchmarks:

 -std=f2018   -g   -O3 

Benchmarks using both C and C++:

 -std=c++17   -std=c18   -g   -O3   -Wno-implicit-function-declaration 

The flags file that was used to format this result can be browsed at
http://www.spec.org/cpu2026/results/flags/llvm-rev-A3.html.

You can also download the XML flags source by saving the following link:
http://www.spec.org/cpu2026/results/flags/llvm-rev-A3.xml.