Figures

Apple M4 GPU: recorded figures
Metric Value Evidence Basis Method Source
Geekbench Compute score, metal gpu 54 482 Confirmed Measured Benchmark chart average Published independently geekbench-compute 6 , metal A dated capture of the benchmark publisher’s rolling average for this graphics processor, read from the 18 July 2026 Metal chart. It may be compared only with another Geekbench 6 Metal figure from this curation round, never with an OpenCL figure, a later version or a processor score. The chart also lists this processor under “Apple M4 GPU”, at 54 196; the row naming the product as its maker does was taken, and the disagreement is recorded in the conflict ledger.
Geekbench Compute score, opencl gpu 32 716 Confirmed Measured Benchmark chart average Published independently geekbench-compute 7 , opencl A dated capture of the benchmark publisher’s rolling average for this graphics processor, read from the 25 July 2026 OpenCL chart. Version 7 calibrates a compute result against a baseline of 100,000, so it may be compared only with another Geekbench 7 OpenCL figure from this curation round, never with a Metal figure, an earlier version or a processor score. The chart also lists this processor under “QUALCOMM Apple M4 GPU730”, at 8 518; the row naming the product as its maker does was taken, and the disagreement is recorded in the conflict ledger.

Conflicting sources

Geekbench Compute score

  1. 54 482 accepted

    Device and Score table, row "Apple M4": 54482.

  2. 54 196

    Device and Score table, row "Apple M4 GPU": 54196.

Editorial decision 2026-08-18

The chart keys its rows on the string a driver reports, so one product reaches it under several names with several averages. The rule this catalog applies, here and in every later round: take the row that names the product as its maker names it, and treat a string carrying tokens the maker does not use — a “GPU” suffix, a die designation, another company’s prefix — as the driver’s name rather than the product’s. The same pair of strings as the M1, decided the same way. Here the two averages agree to within half a percent, which is evidence that the split is a labeling artifact rather than two different things being measured.

Geekbench Compute score

  1. 32 716 accepted

    Device and Score table, row "Apple M4": 32716.

  2. 8518

    Device and Score table, row "QUALCOMM Apple M4 GPU730": 8518.

Editorial decision 2026-08-18

The chart keys its rows on the string a driver reports, so one product reaches it under several names with several averages. The rule this catalog applies, here and in every later round: take the row that names the product as its maker names it, and treat a string carrying tokens the maker does not use — a “GPU” suffix, a die designation, another company’s prefix — as the driver’s name rather than the product’s. Qualcomm makes no Apple silicon, so the prefix is a driver’s misreport and the row cannot be what it says it is. Its average is a quarter of the other, which on its own would not decide anything — the identification does.

Fitted to

Sources

Back to top