Apple Silicon's first generation skip: M6 ships base-only, M6 Pro/Max cut to make room for M7
TL;DR
Bloomberg reports Apple will skip M6 Pro and M6 Max entirely, pushing pro-tier Mac chips to the M7 family in late 2027 — the first break in Apple Silicon's annual Pro/Max cadence since 2020.
Bloomberg's Mark Gurman reported on June 25 that Apple will completely skip M6 Pro and M6 Max, deferring pro-tier workstation chips to the M7 family in late 2027. This is the first break in the «base / Pro / Max / Ultra same-generation» rhythm Apple Silicon has held since M1 in 2020.
The new roadmap: M5 Ultra (new Mac Studio) and a single M6 base SKU ship by end of year; M7 base arrives early 2027; M7 Pro/Max land late 2027; M7 Ultra in 2028. For users on M3 Pro/Max today — VFX, ML engineers, large Xcode builds — the next upgrade window shifts from autumn 2026 to late 2027: nearly two years of waiting.
Gurman's reasoning is AI compute and memory bandwidth. M6 base targets 200 GB/s and 12-core GPU on TSMC 2nm. M7 base pulls bandwidth to 240 GB/s. «The M7 family is redesigned around on-device AI and GPU-heavy workloads,» Gurman writes — i.e. M6 Pro/Max on the original plan would already lag Apple Intelligence's demand curve. Better to free the wafer capacity and jump straight to M7.
The market impact: «annual Pro chip upgrade» has been Apple's most reliable hardware promise for six years (M1 Pro/Max 2021 through M5 Pro/Max 2025). That cadence breaks here. Mac Pro and 16-inch MacBook Pro buyers lose their 2026 refresh.
This connects: Apple raised Mac/iPad prices June 25, Cook says more to come (June 26), Samsung commits $648B for memory expansion the same week, Xbox raises prices for August. Memory and AI compute have rewired consumer electronics — M6 Pro/Max being cut is the quietest but most structural rearrangement.
Upside for TSMC: 2nm capacity originally allocated to M6 Pro/Max merges into the M7 ramp. Upside for NVIDIA: pro ML workloads on the Mac get deferred two years; H-series and Blackwell hold the workstation market.
via Bloomberg / MacRumors
The new roadmap: M5 Ultra (new Mac Studio) and a single M6 base SKU ship by end of year; M7 base arrives early 2027; M7 Pro/Max land late 2027; M7 Ultra in 2028. For users on M3 Pro/Max today — VFX, ML engineers, large Xcode builds — the next upgrade window shifts from autumn 2026 to late 2027: nearly two years of waiting.
Gurman's reasoning is AI compute and memory bandwidth. M6 base targets 200 GB/s and 12-core GPU on TSMC 2nm. M7 base pulls bandwidth to 240 GB/s. «The M7 family is redesigned around on-device AI and GPU-heavy workloads,» Gurman writes — i.e. M6 Pro/Max on the original plan would already lag Apple Intelligence's demand curve. Better to free the wafer capacity and jump straight to M7.
The market impact: «annual Pro chip upgrade» has been Apple's most reliable hardware promise for six years (M1 Pro/Max 2021 through M5 Pro/Max 2025). That cadence breaks here. Mac Pro and 16-inch MacBook Pro buyers lose their 2026 refresh.
This connects: Apple raised Mac/iPad prices June 25, Cook says more to come (June 26), Samsung commits $648B for memory expansion the same week, Xbox raises prices for August. Memory and AI compute have rewired consumer electronics — M6 Pro/Max being cut is the quietest but most structural rearrangement.
Upside for TSMC: 2nm capacity originally allocated to M6 Pro/Max merges into the M7 ramp. Upside for NVIDIA: pro ML workloads on the Mac get deferred two years; H-series and Blackwell hold the workstation market.
via Bloomberg / MacRumors
