Apple · M5 · Tablet

Best AI Models for iPad Pro M5 13" (16 GB)

Runs 9B models well; 14B only just fits in 10 GB, and 27–31B doesn't fit.

16 GB unified10 GB for models153 GB/s16 TFLOPS FP162025

Same chip in

iPad Pro 13"MacBook Pro 14-inch$1,839 MacBook Air 13"MacBook Air 15"MacBook Pro 14"

What LLMs can the iPad Pro M5 13" run?

#	Model	Speed	Quant · Memory
1	Qwen3.5 9BAlibabaHow to run → Q4_K_M · 6.6 GB · 16.8 t/s	16.8 tok/s	Q4_K_M · 6.6 GB
2	Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 32.9 t/s	32.9 tok/s	Q4_K_M · 3.4 GB
3	Gemma 2 9B ITGoogle Q4_K_M · 7.3 GB · 17.6 t/s	17.6 tok/s	Q4_K_M · 7.3 GB
4	Gemma 3 4B ITGoogle Q4_K_M · 5.0 GB · 37.7 t/s	37.7 tok/s	Q4_K_M · 5.0 GB
5	Ling 3.0 TinyInclusion AI Q4_K_M · 5.5 GB · 73.9 t/s	73.9 tok/s	Q4_K_M · 5.5 GB

#	Model	Speed	Coding Score	Quant · Memory
1	Llama 3.1 8B InstructMetaHow to run → Q4_K_M · 7.9 GB · 20.2 t/s	20.2 tok/s		Q4_K_M · 7.9 GB
2	Qwen3.5 9BAlibaba Q4_K_M · 6.6 GB · 16.8 t/s	16.8 tok/s	overall	Q4_K_M · 6.6 GB
3	Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 32.9 t/s	32.9 tok/s	overall	Q4_K_M · 3.4 GB
4	Ling 3.0 TinyInclusion AI Q4_K_M · 5.5 GB · 73.9 t/s	73.9 tok/s		Q4_K_M · 5.5 GB
5	Gemma 4 E4B ITGoogle Q4_K_M · 5.6 GB · 20.1 t/s	20.1 tok/s		Q4_K_M · 5.6 GB

#	Model	Speed	Quant · Memory
1	Qwen3.5 9BAlibabaHow to run → Q4_K_M · 6.6 GB · 16.8 t/s	16.8 tok/s	Q4_K_M · 6.6 GB
2	Qwen3 8BAlibaba Q4_K_M · 7.6 GB · 19.4 t/s	19.4 tok/s	Q4_K_M · 7.6 GB
3	Llama 3.1 8B InstructMeta Q4_K_M · 7.9 GB · 20.2 t/s	20.2 tok/s	Q4_K_M · 7.9 GB
4	Gemma 3 4B ITGoogle Q4_K_M · 5.0 GB · 37.7 t/s	37.7 tok/s	Q4_K_M · 5.0 GB
5	Ling 3.0 TinyInclusion AI Q4_K_M · 5.5 GB · 73.9 t/s	73.9 tok/s	Q4_K_M · 5.5 GB

#	Model	Speed	Quant · Memory
1	Qwen3.5 9BAlibabaHow to run → Q4_K_M · 6.6 GB · 16.8 t/s	16.8 tok/s	Q4_K_M · 6.6 GB
2	Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 32.9 t/s	32.9 tok/s	Q4_K_M · 3.4 GB
3	Llama 3.1 8B InstructMeta Q4_K_M · 7.9 GB · 20.2 t/s	20.2 tok/s	Q4_K_M · 7.9 GB
4	Ling 3.0 TinyInclusion AI Q4_K_M · 5.5 GB · 73.9 t/s	73.9 tok/s	Q4_K_M · 5.5 GB
5	Gemma 4 E4B ITGoogle Q4_K_M · 5.6 GB · 20.1 t/s	20.1 tok/s	Q4_K_M · 5.6 GB

#	Model	Speed ↓	Quant · Memory
1	Qwen3.5 4BAlibabaHow to run → Q4_K_M · 3.4 GB · 32.9 t/s	32.9 tok/s	Q4_K_M · 3.4 GB
2	Ling 3.0 TinyInclusion AI Q4_K_M · 5.5 GB · 73.9 t/s	73.9 tok/s	Q4_K_M · 5.5 GB
3	Gemma 4 E4B ITGoogle Q4_K_M · 5.6 GB · 20.1 t/s	20.1 tok/s	Q4_K_M · 5.6 GB
4	Qwen3 8BAlibaba Q4_K_M · 7.6 GB · 19.4 t/s	19.4 tok/s	Q4_K_M · 7.6 GB
5	Qwen3.5 9BAlibaba Q4_K_M · 6.6 GB · 16.8 t/s	16.8 tok/s	Q4_K_M · 6.6 GB

#	Model	Speed	Quant · Memory
1	Qwen3.5 9BAlibabaHow to run → Q4_K_M · 6.6 GB · 16.8 t/s	16.8 tok/s	Q4_K_M · 6.6 GB
2	Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 32.9 t/s	32.9 tok/s	Q4_K_M · 3.4 GB
3	Gemma 3 4B ITGoogle Q4_K_M · 5.0 GB · 37.7 t/s	37.7 tok/s	Q4_K_M · 5.0 GB
4	MiMo V2.6 Distill Qwen 9BXiaomi Q4_K_M · 6.5 GB · 17.2 t/s	17.2 tok/s	Q4_K_M · 6.5 GB
5	MiniCPM V 4.7 1BOpenBMB Q4_K_M · 1.6 GB · 83.7 t/s	83.7 tok/s	Q4_K_M · 1.6 GB

See all 244 models on the iPad Pro M5 13" →Bars = traction for models without a Score yet

About the iPad Pro M5 13" (16 GB)

The iPad Pro M5 13" is a mid-range 16 GB tablet: it runs 9B models well. Among 16 GB machines, only the M1 Pro and M2 Pro write faster. iPadOS keeps about 6 GB, leaving 10 GB for models. With 16 TFLOPS it reads long prompts faster than most 16 GB machines: a 4K-token prompt takes 25 s. M5 generation, released 2025.

Speed in 16 GB#3153 GB/s

Prompt reading#116 TFLOPS

For models10 GBiPadOS keeps 6 GB

ChipM5released 2025

iPad Pro M5 13" strengths and weaknesses

Strengths

Runs 7–9B models: Qwen3.5 9B at 16.8 tok/s in 6.6 GB

Weaknesses

27–31B models don't fit: Qwen3.8 27B needs 17.9 GB at Q4 vs 10 GB usable here
Little room for context: Qwen3 14B uses 9.6 of 10 GB usable, 0.4 GB left
Slow prompt reading: A 32K-token prompt on Qwen3.5 9B waits 3.3 min for the first token
No CUDA: vLLM, AWQ, GPTQ and FP8 builds don't run; use GGUF or MLX

Memory for models

10 GB of 16 GB

9B6.6 GB14B9.6 GB

For models 10 GBiPadOS keeps 6 GBModel size at Q4

27–31B models need 17.9 GB at Q4, more than the 10 GB available.Or rent an RTX 3090 (24 GB) from $0.19/hr →

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.

Machine, 21% sloweriPad Pro M4 13" (16 GB)Apple · 12.5 GB for models · M4

▼ 21% speed▲ 2.5 GBno price

56% faster, 86 GB more memoryAsus ROG Flow Z13 2025Asus · 96 GB for models · Ryzen AI Max+ 395

▲ 56% speed▲ 86 GB$2,799

iPad Pro M5 13" tokens per second by model size

Output speed depends on memory bandwidth (153 GB/s). Grey rows need more than the 10 GB available.

9BQwen3.5 9B · Q4_K_M · 6.6 GB
16.8tok/s
14BQwen3 14B · IQ3_M · 9.6 GB tight · IQ3_M
14.7tok/s
27–31BQwen3.8 27B · not at 4-bit
needs 17.9 GB
35B MoEQwen3.6 35B A3B · not at 4-bit
needs 22.0 GB
70BLlama 3.3 70B Instruct · not at 4-bit
needs 51.3 GB
120B MoEGPT OSS 120B · not at 4-bit
needs 70.8 GB

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

Chip generations: writing vs compute

M1 → M5

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

Writing speed follows memory bandwidth and barely moved: 7.5 to 16.8 tok/s since the M1. Compute, which sets how fast prompts are read, grew 3.1×. Indexed to the M1; speeds on Qwen3.5 9B, same 16 GB.

All 903 models on the iPad Pro M5 13"

Featured only

Best quant that fits in 10 GB, ordered by llmrun Score. Fit grades how well it runs here, S best. How ranking works

LLM models compatible with iPad Pro M5 13" (16 GB) — ranked by Score, recency and Fit
Model	Quant	VRAM	Speed	Context	Fit
Llama3 8B Chinese Chat8.0BChat Q4_K_M·19.9 t/s	Q4_K_M	6.2 GB62%	19.9 t/s	8K	B60, Decent
Qwen3.5 4B MiniFantasy MTP4.7BChatRoleplay FP16·10.9 t/s	FP16	10.0 GB100%	10.9 t/s	262K	D18, Barely runs
Opt 1.3B1.3BChat FP16·37.4 t/s	FP16	2.9 GB29%	37.4 t/s	2K	A74, Runs well
Finance Llama3 8B8.0BChat Q4_K_M·19.9 t/s	Q4_K_M	6.2 GB62%	19.9 t/s	8K	B60, Decent
Qwen3.6 27B27.8BVisionFunctions IQ2_S·11.4 t/s	IQ2_S	9.9 GB99%	11.4 t/s	262K	D20, Barely runs
EuroMoE 2.6B A0.6B 25122.6BChat Q4_K_M·121.7 t/s	Q4_K_M	2.0 GB20%	121.7 t/s	4K	S95, Runs great
NuExtract 1.53.8BChat Q4_K_M·31.5 t/s	Q4_K_M	9.0 GB90%	31.5 t/s	131K	B57, Decent
NuExtract3.8BChat Q4_K_M·31.5 t/s	Q4_K_M	4.2 GB42%	31.5 t/s	4K	A71, Runs well
Openchat 3.5 12107.2BChat Q4_K_M·21.8 t/s	Q4_K_M	5.7 GB57%	21.8 t/s	8K	B62, Decent
Mistral 7B v0.27.2BChat Q4_K_M·21.8 t/s	Q4_K_M	6.8 GB68%	21.8 t/s	33K	B62, Decent
Zephyr 7B Beta7.2BChat Q4_K_M·21.8 t/s	Q4_K_M	6.8 GB68%	21.8 t/s	33K	B62, Decent
Human Like LLama3 8B Instruct8.0BChat Q4_K_M·19.9 t/s	Q4_K_M	6.2 GB62%	19.9 t/s	8K	B60, Decent
Starcoder2 3B3.0BChatCode Q4_K_M·49.1 t/s	Q4_K_M	2.6 GB26%	49.1 t/s	16K	A79, Runs well
Dolphin 2.9 Llama3 8B8.0BChat Q4_K_M·19.9 t/s	Q4_K_M	6.2 GB62%	19.9 t/s	8K	B60, Decent
LFM2 1.2B1.2BChatFunctions Q4_K_M·100.1 t/s	Q4_K_M	1.2 GB12%	100.1 t/s	128K	S91, Runs great

iPad Pro M5 13" specs

Specs

Chip: M5
CPU cores: 10
GPU cores: 10
Unified memory: 16 GB LPDDR5X-9600
For models: 10 GB (iPadOS keeps 6 GB)
Memory bandwidth: 153 GB/s
FP16 (dense): 16 TFLOPS
NPU: 16-core Neural Engine with Neural Accelerators
Size: tablet
Released: 2025-10-22

iPad Pro M5 13" FAQ

How much memory is available for AI on the iPad Pro M5 13" (16 GB)?

About 10 GB of the 16 GB. iPadOS keeps roughly 6 GB for itself and other apps.

What is the largest model the iPad Pro M5 13" (16 GB) can run?

Qwen3.5 9B (9.7B) is the largest at 4-bit or better: at Q4_K_M it uses 6.6 GB, 66% of the available memory, and runs at about 16.8 tok/s.

Can the iPad Pro M5 13" (16 GB) run 27–31B models?

No. Qwen3.8 27B needs 17.9 GB at Q4, more than the 10 GB available.

How long does the iPad Pro M5 13" (16 GB) take to read a long prompt?

On Qwen3.5 9B, about 25 s for a 4K-token prompt and 3.3 min for a 32K-token document before the first word appears.

8 GB GPUs All Macs Tablets Mac Mini M2 Pro MacBook Pro 14" M1 Pro MacBook Pro 14" M2 Pro MacBook Pro 16" M1 Pro MacBook Pro 16" M2 Pro Mac Mini M4 Pro Best coding LLMs All hardware