Apple’s silicon roadmap keeps moving, and the first public Geekbench results for the M5 MacBook Pro point to another clear step. Early numbers show faster single-core, stronger multi-core, and a big jump in Apple’s Metal GPU score, extending the lead Apple built with M1. These are first-wave uploads, so treat them as a snapshot, not the final average. The trend already looks consistent with Apple’s launch messaging about higher performance and efficiency.
Method in brief: Geekbench scores come from user-submitted runs on the Geekbench Browser. The site aggregates multiple results and posts per-model pages that stabilize as more tests arrive. Early spikes usually settle over time. Keep that in mind while reading the figures below.
Don’t miss the best of The Mac Observer
Set us as a preferred source and our Apple reporting ranks higher in your Google Search results and Discover feed — one tap, no account changes.
Baseline: where M1 started
The 13-inch MacBook Pro with M1 posted about 2,319 in single-core and 8,173 in multi-core on Geekbench 6 in widely shared results. That baseline still shapes how many users experience macOS today because M1 laptops remain common in newsrooms, classrooms, and code shops. It set the ground for Apple’s move away from Intel and gave buyers strong single-thread performance along with excellent battery life.
Since then, Apple has added cores, raised clocks, and widened memory bandwidth. Each generation brought modest single-core gains and larger multi-core and GPU gains. The pattern matters because it explains why even older Apple silicon still feels quick for many everyday tasks while creative and compute heavy work scales more with each step.
CPU and GPU performance evolution
This first chart tracks single-core, multi-core, and Metal GPU scores from M1 through M5, normalized to M1 equals 100 percent. The bars form a clear stair-step. Single-core rises steadily. Multi-core climbs faster as Apple expands performance and efficiency cores.
Metal GPU shows the sharpest slope, reflecting architectural updates and higher memory bandwidth. The M5 bars extend that slope and bring the biggest jump on the graphics side, which lines up with Apple’s public focus on GPU and AI throughput.
M4 vs M5, normalized head to head
The second chart isolates the past two generations. Normalizing 2024 to 100 percent makes the change obvious at a sight. Early M5 single-core sits above 110 percent. Multi-core climbs into the high teens over M4. Metal GPU shows the largest relative jump.
This simple view helps a buyer choose between last year’s discounted model and the newest configuration. If you value graphics, the case for M5 strengthens. If your work is light and single-threaded, last year’s machine remains a smart purchase.
IPC progress, not just clocks
The third visual looks at estimated single-core IPC over time. M1 to M3 delivered small but steady steps. M4 accelerated the curve, and M5 pushes it forward again.
IPC growth matters because it improves everyday responsiveness in code compiles, photo edits, and browsing even before you stress all cores. It explains why Apple silicon machines feel quick across generations and why users with M1 laptops still report smooth performance for general work.
How far behind are older machines
The map reframes the story by asking how much faster M5 is than earlier chips. On single-core, M1 trails by close to two times. On multi-core and GPU, the gap widens more. This lens is useful for upgrade math. If you live in Xcode, Metal, Lightroom, or Blender, the multi-core and GPU deltas save real time. If you mostly browse, write, and hop on calls, the single-core curve is the one to watch.
How this stacks up against the PC side
Debates around ARM, x86, and CUDA point to the same tension. Apple’s hardware pace is strong. Software ecosystems decide where pros land. Nvidia’s CUDA remains dominant in research and industry, which still nudges some developers toward Linux workstations.
Windows on ARM is improving and Qualcomm-powered laptops are finally credible for office work, but translation quality and app availability remain uneven next to Apple’s Rosetta approach. ARM keeps gaining ground in laptops, and Apple set the modern template for that shift.
What early buyers should know
If you own an M1 MacBook Pro and handle heavier media, data, or ML tasks, the combined multi-core and GPU uplift to M5 will save time and battery. If your daily work is writing, spreadsheets, and web apps, an M4 or discounted M3 can still feel quick because Apple’s single-core performance has been strong since day one.
For M4 owners, the M5 gain looks meaningful on graphics and double-digit on CPU, but it is not a night and day jump for light workloads. Give the averages time to settle on the Geekbench Browser before you make a final call.
Bottom line
Early M5 results back up Apple’s claims. You get higher single-core, a healthy multi-core lead over M4, and a clear GPU step. The four charts sit under matching subheads so readers can pair data and context. Apple keeps lifting per-core performance while scaling cores and bandwidth, and the M5 generation doubles down on graphics.
If the averages hold as more results arrive, 2025 looks like another easy year to recommend a MacBook Pro to power users and another difficult year for rivals trying to match the pace.
Discussion