Apple · M3 Pro · Laptop

Best AI Models for MacBook Pro 16" M3 Pro (18 GB)

Runs 9B models well; dense 14B fits but is slow at 11 tok/s, and 27–31B only just fit in 14.5 GB.

18 GB unified14.5 GB for models150 GB/s13 TFLOPS FP162023

Memory

Same chip in

What LLMs can the MacBook Pro 16" M3 Pro run?

#	Model	Speed	Quant · Memory
1	Qwen3.5 9BAlibabaHow to run → Q4_K_M · 6.6 GB	16.5 tok/s	Q4_K_M · 6.6 GB
2	GPT OSS 20BOpenAI Q4_K_M · 13.1 GB	33.9 tok/s	Q4_K_M · 13.1 GB
3	Qwen3 8BAlibaba Q4_K_M · 7.6 GB	19.0 tok/s	Q4_K_M · 7.6 GB
4	Gemma 3 4B ITGoogle Q4_K_M · 5.0 GB	37.0 tok/s	Q4_K_M · 5.0 GB
5	Gemma 4 E4B ITGoogle Q4_K_M · 5.6 GB	19.7 tok/s	Q4_K_M · 5.6 GB

#	Model	Speed	Coding Score	Quant · Memory
1	GPT OSS 20BOpenAIHow to run → Q4_K_M · 13.1 GB	33.9 tok/s		Q4_K_M · 13.1 GB
2	Llama 3.1 8B InstructMeta Q4_K_M · 7.9 GB	19.8 tok/s		Q4_K_M · 7.9 GB
3	Qwen3.5 9BAlibaba Q4_K_M · 6.6 GB	16.5 tok/s	overall	Q4_K_M · 6.6 GB
4	Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB	32.2 tok/s		Q4_K_M · 3.4 GB
5	Gemma 4 E4B ITGoogle Q4_K_M · 5.6 GB	19.7 tok/s		Q4_K_M · 5.6 GB

#	Model	Speed	Quant · Memory
1	Qwen3.5 9BAlibabaHow to run → Q4_K_M · 6.6 GB	16.5 tok/s	Q4_K_M · 6.6 GB
2	Llama 3.1 8B InstructMeta Q4_K_M · 7.9 GB	19.8 tok/s	Q4_K_M · 7.9 GB
3	Llama 3.2 1B InstructMeta Q4_K_M · 2.3 GB	128 tok/s	Q4_K_M · 2.3 GB
4	GPT OSS 20BOpenAI Q4_K_M · 13.1 GB	33.9 tok/s	Q4_K_M · 13.1 GB
5	Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB	32.2 tok/s	Q4_K_M · 3.4 GB

#	Model	Speed ↓	Quant · Memory
1	GPT OSS 20BOpenAIHow to run → Q4_K_M · 13.1 GB	33.9 tok/s	Q4_K_M · 13.1 GB
2	Ling 3.0 TinyInclusion AI Q4_K_M · 5.5 GB	73.0 tok/s	Q4_K_M · 5.5 GB
3	Gemma 4 E4B ITGoogle Q4_K_M · 5.6 GB	19.7 tok/s	Q4_K_M · 5.6 GB
4	Qwen3 8BAlibaba Q4_K_M · 7.6 GB	19.0 tok/s	Q4_K_M · 7.6 GB
5	Qwen3.5 9BAlibaba Q4_K_M · 6.6 GB	16.5 tok/s	Q4_K_M · 6.6 GB

#	Model	Speed	Quant · Memory
1	Qwen3.5 9BAlibabaHow to run → Q4_K_M · 6.6 GB	16.5 tok/s	Q4_K_M · 6.6 GB
2	Gemma 3 4B ITGoogle Q4_K_M · 5.0 GB	37.0 tok/s	Q4_K_M · 5.0 GB
3	Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB	32.2 tok/s	Q4_K_M · 3.4 GB
4	MiMo V2.6 Distill Qwen 9BXiaomi Q4_K_M · 6.5 GB	16.9 tok/s	Q4_K_M · 6.5 GB
5	Cosmos Reason2 2BNVIDIA Q4_K_M · 3.5 GB	65.2 tok/s	Q4_K_M · 3.5 GB

See all 274 models on the MacBook Pro 16" M3 Pro →Bars = traction for models without a Score yet
Task rankings use that task's benchmarks, topped up from Overall when short; bars mark well-known models with no Score yet.

About the MacBook Pro 16" M3 Pro (18 GB)

The MacBook Pro 16" M3 Pro runs 9B models well. macOS keeps about 3.5 GB, leaving 14.5 GB for models. Prompt reading is its weak spot: a 4K-token prompt waits 31 s before the first word on Qwen3.5 9B. M3 Pro generation, released 2023; the newer M4 Pro writes 82% faster with 1.3× the compute.

For models14.5 GBmacOS keeps 3.5 GB

ChipM3 Proreleased 2023

MacBook Pro 16" M3 Pro strengths and weaknesses

Strengths

Runs 7–9B models: Qwen3.5 9B at 16.5 tok/s in 6.6 GB

Weaknesses

35B MoE models don't fit: Qwen3.6 35B A3B needs 22 GB at Q4 vs 14.5 GB usable here
Slow on large dense models: Qwen3.8 27B at 7.9 tok/s
Slow prompt reading: A 32K-token prompt on Qwen3.5 9B waits 4.1 min for the first token
No CUDA: vLLM, AWQ, GPTQ and FP8 builds don't run; use GGUF or MLX

Memory for models

14.5 GB of 18 GB

9B6.6 GB14B11.9 GB27–31B13.7 GB

9B6.6 GB27–31B13.7 GB

For models 14.5 GBmacOS keeps 3.5 GBModel size at Q4

35B MoE models need 22 GB at Q4, more than the 14.5 GB available. The 36 GB config fits Qwen3.8 27B →

Similar devices and alternatives

Other memory sizes of the same machine first, then machines and GPUs people weigh against it, each with the one reason it is worth a look. Speeds on Qwen3.5 9B.

Same chip, 2× the memoryMacBook Pro 16" M3 Pro (36 GB)Apple · 32.5 GB for models · M3 Pro

same speed▲ 18 GBno price

Same family Newer machine, 20% slowerMac Mini M4 (16 GB)Apple · 12.5 GB for models · M4

▼ 20% speed▼ 2 GB$799

82% faster, 6 GB more memoryMac Mini M4 Pro (24 GB)Apple · 20.5 GB for models · M4 Pro

▲ 82% speed▲ 6 GB$1,599

173% faster, 18 GB more memoryMac Studio M4 Max (36 GB)Apple · 32.5 GB for models · M4 Max

▲ 173% speed▲ 18 GB$2,499

Reads prompts 29× faster, 24 GB for modelsRTX 4090NVIDIA · 24 GB VRAM · GPU

▲ 29× prompts▲ 524% speed▲ 9.5 GB$1,599rent $0.34/hr

Reads prompts 37× faster, 32 GB for modelsRTX 5090NVIDIA · 32 GB VRAM · GPU

▲ 37× prompts▲ 1010% speed▲ 17.5 GB$5,376rent $0.59/hr

MacBook Pro 16" M3 Pro tokens per second by model size

Output speed depends on memory bandwidth (150 GB/s). Grey rows need more than the 14.5 GB available.

9BQwen3.5 9B · Q4_K_M · 6.6 GB
16.5tok/s
14BQwen3 14B · Q4_K_M · 11.9 GB
11.1tok/s
27–31BQwen3.8 27B · IQ3_M · 13.7 GB tight · IQ3_M
7.9tok/s
35B MoEQwen3.6 35B A3B · does not fit
needs 22.0 GB
70BLlama 3.3 70B Instruct · does not fit
needs 51.3 GB
120B MoEGPT OSS 120B · does not fit
needs 70.8 GB

Estimated speeds: Q4_K_M, 4K context, one user. How we estimate

Pro generations: writing vs compute

M1 Pro → M5 Pro

Writing speed (tok/s)Compute (TFLOPS FP16)

Writing speed follows memory bandwidth and barely moved: 22.0 to 33.8 tok/s since the M1 Pro. Compute, which sets how fast prompts are read, grew 3.1×. Indexed to the M1 Pro; speeds on Qwen3.5 9B, same 16 GB.

All 975 models on the MacBook Pro 16" M3 Pro

Featured only

Best quant that fits in 14.5 GB, ordered by llmrun Score. Fit grades how well it runs here, S best. How ranking works

LLM models compatible with MacBook Pro 16" M3 Pro (18 GB) — ranked by Score, recency and Fit
Model	Quant	VRAM	Speed	Context	Fit
Kappa 20B 131k20.9BChat 20.9B·Q4_K_M·33.9 t/s·131K ctx	Q4_K_M	13.1 GB91%	33.9 t/s	131K	B57, Decent
SpatialLM1.1 Qwen 0.5B604MChat 604M·BF16·68.6 t/s·33K ctx	BF16	1.7 GB12%	68.6 t/s	33K	S85, Runs great
Qwen3 1.7B Abliterated1.7BChat 1.7B·Q4_K_M·93.7 t/s	Q4_K_M	2.8 GB19%	93.7 t/s	—	S90, Runs great
ReaderLM v21.5BChat 1.5B·Q4_K_M·82.0 t/s·513K ctx	Q4_K_M	1.7 GB12%	82.0 t/s	513K	S88, Runs great
Nemotron Research Reasoning Qwen 1.5B1.8BChatReasoning 1.8B·Q4_K_M·73.9 t/s·131K ctx	Q4_K_M	1.8 GB13%	73.9 t/s	131K	S86, Runs great
DeepSeek Coder v2 Lite Instruct15.7BChatCode 15.7B·Q4_K_M·45.8 t/s·164K ctx	Q4_K_M	10.2 GB71%	45.8 t/s	164K	A78, Runs well
TLive Omni 4B5.8BChatVisionFunctions 5.8B·BF16·8.7 t/s·262K ctx	BF16	13.3 GB92%	8.7 t/s	262K	D29, Barely runs
Arch Router 1.5B1.5BChat 1.5B·Q4_K_M·82.0 t/s·33K ctx	Q4_K_M	1.7 GB12%	82.0 t/s	33K	S88, Runs great
Meme Trix MoE 14B A8B V113.7BChat 13.7B·Q4_K_M·18.1 t/s·1073K ctx	Q4_K_M	10.7 GB73%	18.1 t/s	1073K	B58, Decent
MediPhi Instruct3.8BChat 3.8B·Q4_K_M·30.9 t/s·131K ctx	Q4_K_M	9.0 GB62%	30.9 t/s	131K	A71, Runs well
OpenMath Nemotron 1.5B1.5BChatMath 1.5B·Q4_K_M·82.0 t/s·131K ctx	Q4_K_M	1.7 GB12%	82.0 t/s	131K	S88, Runs great
Tiny Dummy Qwen21MChat 1M·Q4_K_M·350.0 t/s·33K ctx	Q4_K_M	0.5 GB4%	350.0 t/s	33K	S100, Runs great
Racka 4B4.0BChat 4.0B·Q4_K_M·39.6 t/s	Q4_K_M	4.8 GB33%	39.6 t/s	—	A75, Runs well
Thinkless 1.5B RL DeepScaleR1.8BChat 1.8B·Q4_K_M·73.9 t/s·33K ctx	Q4_K_M	1.8 GB13%	73.9 t/s	33K	S86, Runs great
Qwen3.5 4B MiniFantasy MTP4.7BChatRoleplay 4.7B·FP16·10.7 t/s·262K ctx	FP16	10.0 GB69%	10.7 t/s	262K	B49, Decent

MacBook Pro 16" M3 Pro specs

Specs

Chip: M3 Pro
CPU cores: 12
GPU cores: 18
Unified memory: 18 GB
For models: 14.5 GB (macOS keeps 3.5 GB)
Memory bandwidth: 150 GB/s
FP16 (dense): 12.9 TFLOPS
Neural Engine: 18 TOPS
Released: 2023-11-07

MacBook Pro 16" M3 Pro FAQ

How much memory is available for AI on the MacBook Pro 16" M3 Pro (18 GB)?

About 14.5 GB of the 18 GB. macOS keeps roughly 3.5 GB for itself and other apps.

What is the largest model the MacBook Pro 16" M3 Pro (18 GB) can run?

GPT OSS 20B (20.9B) at Q4_K_M uses 13.1 GB, 91% of the available memory, and runs at about 33.9 tok/s.

Can the MacBook Pro 16" M3 Pro (18 GB) run Qwen3.5 9B?

Yes. Qwen3.5 9B fits at Q4_K_M, using 6.6 GB of the 14.5 GB available, and it is the best model that runs well on this machine.

Can the MacBook Pro 16" M3 Pro (18 GB) run 35B MoE models?

No. Qwen3.6 35B A3B needs 22.0 GB at Q4, more than the 14.5 GB available.

How long does the MacBook Pro 16" M3 Pro (18 GB) take to read a long prompt?

On Qwen3.5 9B, about 31 s for a 4K-token prompt and 4.1 min for a 32K-token document before the first word appears.

Is renting a GPU cheaper than buying the MacBook Pro 16" M3 Pro?

A Tesla V100 (16 GB) rents from $0.10/hr.

12 GB GPUs All Macs Laptops Mac Studio M5 Max MacBook Pro 14" M5 Max MacBook Pro 16" M5 Max MacBook Pro 14" M4 Max Best coding LLMs All hardware