Apple M4 Pro (16-core GPU, Mac mini, 24GB)
Compare Apple M4 Pro (16-core GPU, Mac mini, 24GB) against another machine in the Arena
Reference: Apple M4 on Wikipedia
- Class
- Mac
- Memory
- 24 GB LPDDR5X
- Runs (Q4_K_M)
- —
- Bandwidth
- 273 GB/s
- Released
- 2024-11-08
- Price (US)
- $1599 new as of 2026-08
Sourcing and disambiguation notes
Fills the gap between the base Mac mini (120 GB/s) and the Max-tier machines (410 GB/s and up), which is where most people shopping for a Mac to run models locally actually end up. Apple states 273 GB/s for BOTH M4 Pro GPU bins, 16-core and 20-core -- unlike the M4 Max, where the 32-core bin is 410 GB/s and the 40-core is 546 GB/s, so the extra cores buy prefill throughput here, not decode. No measured record yet: the two llama.cpp runs found on M4 Pro silicon are a 20-core Mac mini and a 16-core MacBook Pro, neither of which is this configuration, and filing them here would attribute a number to a machine that did not produce it.
Specification sources
- https://support.apple.com/en-us/121553 — vendor, 2024-11-08
- https://www.apple.com/mac-mini/specs/ — vendor, 2026-08-23
- https://www.apple.com/newsroom/2024/10/apple-introduces-m4-pro-and-m4-max/ — vendor, 2024-10-30
Measurements
Decode is token generation — the speed you feel while an answer streams. Prefill is prompt processing — the wait before it starts. Why bandwidth predicts decode speed.
No records yet for this hardware. Know of a published benchmark on it? Submit the link.
What verified, single-source and estimated mean, and the same rows with every filter and sort in the benchmarks explorer.
Speed over time
What it can run
Fit is arithmetic, not a measurement — how it is computed. Against 24 GB; models that fit are listed largest first.
No modeled quant fits in 24 GB.