Recommended Free Tools
Some links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.
Geekbench 6.7, released on April 7, 2026, added detection for Intel’s Binary Optimization Tool (BOT). When Geekbench detects BOT during a CPU test, the resulting submission is marked invalid in the Geekbench Browser. That does not mean every Intel score is invalid, that affected records were necessarily deleted, or that Geekbench blocks BOT from running.
What Intel BOT changes
Intel’s Binary Optimization Tool modifies executable instruction sequences at runtime or launch so supported applications can use Intel-specific optimized code paths. Primate Labs reported that BOT supported a limited set of applications, including Geekbench 6, and that its public documentation did not fully describe the transformations involved.
Geekbench’s investigation indicated that BOT appeared to identify supported benchmark executables through a checksum. In practice, that means changing the benchmark binary—for example, moving from one Geekbench 6 release to another—can affect whether the optimization is applied. Primate Labs’ findings are detailed in its initial BOT report and technical investigation.
Why this creates a comparability problem
The issue is not simply that an Intel processor may receive a performance boost. A BOT-enabled run can execute a substantially different instruction stream from a BOT-disabled run, and that transformed code may be tuned specifically for Intel hardware. If AMD, Arm, or another platform runs the ordinary benchmark binary, the two scores are no longer measuring exactly the same workload.
#1 Best Overall
![Express Rip Free CD Ripper Software - Extract Audio in Perfect Digital Quality [PC Download]](https://m.media-amazon.com/images/I/41xx28xHa+L._SL160_.jpg)
- Perfect quality CD digital audio extraction (ripping)
- Fastest CD Ripper available
- Extract audio from CDs to wav or Mp3
- Extract many other file formats including wma, m4q, aac, aiff, cda and more
- Extract many other file formats including wma, m4q, aac, aiff, cda and more
That does not make the performance gain necessarily fake, and it does not establish that Intel’s optimization is fraudulent. Software vendors routinely optimize applications for particular processors. The problem for a cross-platform benchmark database is that selectively transformed benchmark code can make one vendor’s result less comparable with results produced elsewhere.
How large was the reported performance difference?
Primate Labs’ initial March 24 warning said BOT could raise some Geekbench workloads by as much as 40% and overall scores by as much as 8%. Those were initial observed upper bounds, not a universal uplift for every Intel system.
In a later controlled test using an MSI Prestige 16 AI+ with an Intel Core 9 386H and Geekbench 6.3, Geekbench reported the following:
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
| Test | BOT disabled | BOT enabled | Difference |
|---|---|---|---|
| Single-core | 2,955 | 3,119 | +5.5% |
| Multi-core | 16,786 | 17,705 | +5.5% |
Some individual workloads behaved very differently. Geekbench reported gains of up to approximately 30% in workloads including Object Remover and HDR. The average score therefore understates the effect on particular parts of the test.
Rank #2
![DeskFX Free Audio Effects & Audio Enhancer Software [PC Download]](https://m.media-amazon.com/images/I/41fXbDohyuS._SL160_.jpg)
- Transform audio playing via your speakers and headphones
- Improve sound quality by adjusting it with effects
- Take control over the sound playing through audio hardware
In its HDR analysis, Primate Labs reported that total instructions fell from 1.26 trillion without BOT to 1.08 trillion with BOT. Scalar instructions fell from 220 billion to 84.6 billion, while vector instructions rose from 1.25 billion to 18.3 billion. Geekbench interpreted that pattern as evidence that BOT was vectorizing parts of the workload rather than merely rearranging instructions. These figures come from Geekbench’s own testing and should be understood as measurements from its test system, not as an independently replicated benchmark.
What Geekbench 6.7 actually does
Geekbench 6.7 is primarily a detection and labeling update:
- It detects whether Intel BOT is enabled on the system.
- It marks a detected BOT-assisted submission as invalid in the Geekbench Browser.
- It preserves score comparability with Geekbench 6.3, 6.4, 6.5, and 6.6 according to Primate Labs.
- It improves Android SoC identification by reporting manufacturer and model information.
- It improves RISC-V CPU identification.
- It fixes hangs affecting multithreaded workloads on Linux Arm systems.
The release does not say that Geekbench prevents BOT from operating or disables Intel’s feature outside the benchmark. A local test can still complete and display a score even if its uploaded result is later labeled invalid. “The benchmark ran successfully” and “the result is suitable for cross-platform comparison” are separate claims.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →See the official Geekbench 6.7 announcement for the release details.
Rank #3
![WavePad Audio Editing Software - Professional Audio and Music Editor for Anyone [Download]](https://m.media-amazon.com/images/I/B1fcLEGCs6S.png._SL160_.png)
- Full-featured professional audio and music editor that lets you record and edit music, voice and other audio recordings
- Add effects like echo, amplification, noise reduction, normalize, equalizer, envelope, reverb, echo, reverse and more
- Supports all popular audio formats including, wav, mp3, vox, gsm, wma, real audio, au, aif, flac, ogg and more
- Sound editing functions include cut, copy, paste, delete, insert, silence, auto-trim and more
- Integrated VST plugin support gives professionals access to thousands of additional tools and effects
Does “invalid” mean the result was deleted?
Not necessarily. Primate Labs’ wording says detected results are flagged as invalid on the Geekbench Browser. That describes their status; it does not establish that every affected record was erased, removed from the database, or made impossible to view.
Likewise, Geekbench 6.7 does not invalidate every Intel CPU submission. The invalid status applies when BOT is detected. An Intel system running without BOT is not automatically equivalent to a BOT-enabled run, but it is also not automatically invalid merely because it uses an Intel processor.
What happens to older Geekbench results?
Older results are not automatically useless, but they require more caution. Geekbench said Geekbench 6.6 and earlier Windows results on affected systems would continue to receive warnings because those versions could not determine directly whether BOT was active.
Free tools Windows power users keep installed
One-click scans. No signup required.
A historical score should therefore be read alongside its:
Rank #4
![MixPad Multitrack Recording Software for Sound Mixing and Music Production Free [Mac Download]](https://m.media-amazon.com/images/I/51kaO82jYOL._SL160_.jpg)
- Mix an audio, music and voice tracks
- Record single or multiple tracks simultaneously
- Intuitive tools to split, trim, join, and many other editing features
- Loaded with audio effects including EQ, compression, reverb, and more.
- Load an audio file and export to all popular audio formats from studio quality wav to high compression formats
- Geekbench version and minor release;
- operating system;
- Intel CPU model and possible BOT support;
- Browser warning or invalid label;
- hardware configuration and power profile;
- cooling and sustained-clock conditions; and
- source, especially whether it came from a documented review or an anonymous upload.
Geekbench says 6.7 scores remain comparable with versions 6.3 through 6.6, but that does not remove the need to check whether a particular result carries a warning or invalid status. Cross-version compatibility is not the same as proof that every older run used an unmodified executable.
Why Geekbench 6.7 did not show the same uplift in testing
In Primate Labs’ investigation, the tested Geekbench 6.7 installation showed approximately unchanged single-core results and only about a 0.9% multi-core difference when BOT was enabled:
| Test | BOT disabled | BOT enabled | Difference |
|---|---|---|---|
| Single-core | 2,938 | 2,937 | Approximately 0% |
| Multi-core | 16,892 | 17,045 | Approximately +0.9% |
The likely explanation was that the BOT version available during testing had optimized Geekbench 6.3 but not yet Geekbench 6.7. That is a version-specific observation, not proof that BOT could never support or optimize a later Geekbench binary.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errorsGeekbench also measured startup overhead under BOT: approximately 40 seconds on the first Geekbench 6.3 run and two seconds on later runs. Geekbench 6.7 runs took approximately two seconds with BOT enabled in the reported testing, while the delay disappeared when BOT was disabled. The startup delay is a separate behavior; it is not the reason Geekbench labels the result invalid.
Best Value
![Free Fling File Transfer Software for Windows [PC Download]](https://m.media-amazon.com/images/I/41Vq6ZqHfjL._SL160_.jpg)
- Intuitive interface of a conventional FTP client
- Easy and Reliable FTP Site Maintenance.
- FTP Automation and Synchronization
Timeline of the BOT issue
- March 24, 2026: Primate Labs warned that Intel BOT could modify Geekbench 6 and temporarily displayed warnings on results from CPUs that supported BOT.
- March 31, 2026: Geekbench published its technical investigation, including the reported 5.5% overall uplift on its Core 9 386H test system and larger workload-level differences.
- April 7, 2026: Geekbench 6.7 added BOT detection and invalid-result labeling.
- April 28, 2026: Geekbench 6.7.1 fixed an Android and Linux Browser connection problem caused by an outdated bundled LibreSSL version after Cloudflare protection was deployed. Windows, macOS, and iOS users were reported as unaffected. See the Geekbench 6.7.1 announcement.
- July 23, 2026: Geekbench 7 became the newer major release. Geekbench 6.7 remains relevant for interpreting existing Geekbench 6 data and the BOT dispute.
How to evaluate an Intel Geekbench result
For benchmark readers and CPU buyers
- Check the exact Geekbench major and minor version.
- Confirm the CPU model and operating system.
- Look for a Browser warning or an “invalid” label.
- Do not compare a BOT-enabled or possibly BOT-enabled Intel run with an unmodified AMD or Arm run.
- Treat scores from Geekbench 6.6 or older on a BOT-capable Windows system as provisional.
- Be cautious with unusually high scores, engineering samples, unreleased platforms, and anonymous uploads.
- Use application benchmarks, games, encoding, compilation, and other workloads before drawing conclusions about overall CPU performance.
For reviewers and benchmark database users
Use the same Geekbench version across systems, update to a version that can detect BOT, and document the benchmark version and platform details in every chart. Confirm that BOT is disabled where possible, or clearly identify and exclude any result that Geekbench marks invalid. Pair Geekbench with real application tests because neither a BOT-assisted nor an ordinary synthetic score is a complete model of everyday performance.
The practical verdict
Geekbench 6.7 improves the integrity of Intel CPU comparisons by identifying a condition that older versions could not reliably detect. Its key action is not to block Intel BOT or erase every questionable score; it is to mark submissions where BOT is detected as invalid in the Geekbench Browser.
The fairest interpretation is that Intel BOT may be a legitimate optimization technology, but a benchmark result produced with an Intel-specific transformed binary should not be treated as directly comparable with a result produced from the ordinary benchmark code. When reading Intel leaks or historical Geekbench submissions, the benchmark version, Browser status, software environment, and workload context matter as much as the headline score.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
