Apple M4 (10-core GPU, 120 GB/s): edge AI benchmarks

Name: Apple M4 (10-core GPU, 120 GB/s)
Brand: Apple

Apple Mac mini (M4, 2024), small silver aluminium desktop, front view — Apple Mac mini (M4, 2024) — the M4 Pro model shares this enclosure. Photo: Seasider53, CC BY 4.0, via Wikimedia Commons.

Vendor AppleForm factor laptop/desktop SoCCompute GPUIndicative price unknownPower envelope unknownPublished rows 33

CPU-only marks 14 result(s) run on the CPU without the accelerator (per the source's submission notes).

object detection357 FPSB perf/W n/a2026 (year only)

instance segmentation196 FPSB perf/W n/a2026 (year only)

pose estimation370 FPSB perf/W n/a2026 (year only)

image classification417 FPSB perf/W n/a2026 (year only)

Best published result per workload, across model sizes — open “Show data” for like-for-like groups.

Show data — per-workload tables

object detection

Like-for-like group	Best result	Model · precision · stack	Grade	Power	Source · date	Rank in group
nano (YOLO-n) · b=1	357 FPS	YOLO26n FP16 · Ultralytics 8.4.168 Core AI (coreai-torch 0.4.3), macOS 27.0	B	unknown power n/a	Ultralytics docs: Apple Core AI integration benchmark 2026 (year only)	7 of 16

instance segmentation

Like-for-like group	Best result	Model · precision · stack	Grade	Power	Source · date	Rank in group
nano (YOLO-n) · b=1	196 FPS	YOLO26n-seg FP16 · Ultralytics 8.4.168 Core AI (coreai-torch 0.4.3), macOS 27.0	B	unknown power n/a	Ultralytics docs: Apple Core AI integration benchmark 2026 (year only)	3 of 5

pose estimation

Like-for-like group	Best result	Model · precision · stack	Grade	Power	Source · date	Rank in group
nano (YOLO-n) · b=1	370 FPS	YOLO26n-pose FP16 · Ultralytics 8.4.168 Core AI (coreai-torch 0.4.3), macOS 27.0	B	unknown power n/a	Ultralytics docs: Apple Core AI integration benchmark 2026 (year only)	1 of 1

image classification

Like-for-like group	Best result	Model · precision · stack	Grade	Power	Source · date	Rank in group
nano (YOLO-n) · b=1	417 FPS	YOLO26n-cls FP16 · Ultralytics 8.4.168 Core AI (coreai-torch 0.4.3), macOS 27.0	B	unknown power n/a	Ultralytics docs: Apple Core AI integration benchmark 2026 (year only)	2 of 4

semantic segmentation

Like-for-like group	Best result	Model · precision · stack	Grade	Power	Source · date	Rank in group
nano (YOLO-n) · b=1	53.48 FPS	YOLO26n-sem FP16 · Ultralytics 8.4.168 Core AI (coreai-torch 0.4.3), macOS 27.0	B	unknown power n/a	Ultralytics docs: Apple Core AI integration benchmark 2026 (year only)	1 of 3

speech-to-text

◌

No published speech-to-text results for Apple M4 (10-core GPU, 120 GB/s) yet.
We only show cited public figures — nothing is estimated to fill gaps. Requests tell us what to source or test next.

LLM generation

Like-for-like group	Best result	Model · precision · stack	Grade	Power	Source · date	Rank in group
LLM 5–14B · b=1	24.11 tok/s	Llama 2 7B Q4_0 · llama.cpp Metal (commit 8e672ef)	C	unknown power n/a	llama.cpp discussion #4167 scoreboard (via GitHub API) 2026-10-01	38 of 53
LLM ≤4B · b=1	41.31 tok/s	Llama 3.2:3b as in linked issue (Ollama default tags are 4-bit) · Ollama / llama.cpp (see linked issue)	C	30.10 W measured	Jeff Geerling ai-benchmarks README (Findings tables) undated	44 of 64

Compare with another platform (99)

How to read these numbers

Every number is labelled. Confidence: A MLPerf closed, audited · B documented third-party method · C community / review · D vendor-published, not independently verified · E estimate. Power basis: measured with the result · spec vendor spec/envelope · power n/a not published.

Machine-readable: /find.json (see llms.txt); full tables are in “Show data”. Confidence grades: A audited MLPerf closed · B documented third-party · C community · D vendor-published · E estimate. Unknown values are shown as unknown, never estimated. Generated 2026-10-16 from public sources.