Apple · M4 · Laptop

Best AI Models for MacBook Air 13" M4 (24 GB)

Runs 27–31B models at 5 tok/s; 9B and 14B fit but are slow (13 and 9 tok/s), and 70B doesn't fit in 20.5 GB.

24 GB unified20.5 GB for models120 GB/s9 TFLOPS FP162025

Buy$1,399

Memory

16 GB$999 24 GB$1,399 32 GB

Same chip in

MacBook Air 13"$1,399 Mac MiniMacBook Air 15"MacBook Pro 14"iMac 24"iPad Pro 13"

What LLMs can the MacBook Air 13" M4 run?

#	Model	Speed	Quant · Memory
1	Gemma 4 26B A4B ITGoogleHow to run → Q4_K_M · 16.5 GB · 25.8 t/s	25.8 tok/s	Q4_K_M · 16.5 GB
2	Qwen3.5 9BAlibaba Q4_K_M · 6.6 GB · 13.2 t/s	13.2 tok/s	Q4_K_M · 6.6 GB
3	Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 25.8 t/s	25.8 tok/s	Q4_K_M · 3.4 GB
4	GPT OSS 20BOpenAI Q4_K_M · 13.1 GB · 28.1 t/s	28.1 tok/s	Q4_K_M · 13.1 GB
5	LFM2.5 24B A2BLiquid AI Q4_K_M · 15.0 GB · 38.3 t/s	38.3 tok/s	Q4_K_M · 15.0 GB

#	Model	Speed	Coding Score	Quant · Memory
1	Gemma 4 26B A4B ITGoogleHow to run → Q4_K_M · 16.5 GB · 25.8 t/s	25.8 tok/s		Q4_K_M · 16.5 GB
2	GPT OSS 20BOpenAI Q4_K_M · 13.1 GB · 28.1 t/s	28.1 tok/s		Q4_K_M · 13.1 GB
3	Llama 3.1 8B InstructMeta Q4_K_M · 7.9 GB · 15.8 t/s	15.8 tok/s		Q4_K_M · 7.9 GB
4	Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 25.8 t/s	25.8 tok/s	overall	Q4_K_M · 3.4 GB
5	LFM2.5 24B A2BLiquid AI Q4_K_M · 15.0 GB · 38.3 t/s	38.3 tok/s		Q4_K_M · 15.0 GB

#	Model	Speed	Quant · Memory
1	Gemma 4 26B A4B ITGoogleHow to run → Q4_K_M · 16.5 GB · 25.8 t/s	25.8 tok/s	Q4_K_M · 16.5 GB
2	GPT OSS 20BOpenAI Q4_K_M · 13.1 GB · 28.1 t/s	28.1 tok/s	Q4_K_M · 13.1 GB
3	Qwen3 8BAlibaba Q4_K_M · 7.6 GB · 15.2 t/s	15.2 tok/s	Q4_K_M · 7.6 GB
4	Llama 3.1 8B InstructMeta Q4_K_M · 7.9 GB · 15.8 t/s	15.8 tok/s	Q4_K_M · 7.9 GB
5	LFM2.5 24B A2BLiquid AI Q4_K_M · 15.0 GB · 38.3 t/s	38.3 tok/s	Q4_K_M · 15.0 GB

#	Model	Speed	Quant · Memory
1	Qwen3.5 4BAlibabaHow to run → Q4_K_M · 3.4 GB · 25.8 t/s	25.8 tok/s	Q4_K_M · 3.4 GB
2	GPT OSS 20BOpenAI Q4_K_M · 13.1 GB · 28.1 t/s	28.1 tok/s	Q4_K_M · 13.1 GB
3	Llama 3.1 8B InstructMeta Q4_K_M · 7.9 GB · 15.8 t/s	15.8 tok/s	Q4_K_M · 7.9 GB
4	Gemma 4 26B A4B ITGoogle Q4_K_M · 16.5 GB · 25.8 t/s	25.8 tok/s	Q4_K_M · 16.5 GB
5	LFM2.5 24B A2BLiquid AI Q4_K_M · 15.0 GB · 38.3 t/s	38.3 tok/s	Q4_K_M · 15.0 GB

#	Model	Speed ↓	Quant · Memory
1	GPT OSS 20BOpenAIHow to run → Q4_K_M · 13.1 GB · 28.1 t/s	28.1 tok/s	Q4_K_M · 13.1 GB
2	LFM2.5 24B A2BLiquid AI Q4_K_M · 15.0 GB · 38.3 t/s	38.3 tok/s	Q4_K_M · 15.0 GB
3	Gemma 4 26B A4B ITGoogle Q4_K_M · 16.5 GB · 25.8 t/s	25.8 tok/s	Q4_K_M · 16.5 GB
4	Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 25.8 t/s	25.8 tok/s	Q4_K_M · 3.4 GB
5	Qwen3 8BAlibaba Q4_K_M · 7.6 GB · 15.2 t/s	15.2 tok/s	Q4_K_M · 7.6 GB

#	Model	Speed	Quant · Memory
1	Gemma 4 26B A4B ITGoogleHow to run → Q4_K_M · 16.5 GB · 25.8 t/s	25.8 tok/s	Q4_K_M · 16.5 GB
2	Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 25.8 t/s	25.8 tok/s	Q4_K_M · 3.4 GB
3	Gemma 3 4B ITGoogle Q4_K_M · 5.0 GB · 29.6 t/s	29.6 tok/s	Q4_K_M · 5.0 GB
4	Qwen3.5 9BAlibaba Q4_K_M · 6.6 GB · 13.2 t/s	13.2 tok/s	Q4_K_M · 6.6 GB
5	MiniCPM V 4.7 1BOpenBMB Q4_K_M · 1.6 GB · 65.6 t/s	65.6 tok/s	Q4_K_M · 1.6 GB

See all 320 models on the MacBook Air 13" M4 →Bars = traction for models without a Score yet

About the MacBook Air 13" M4 (24 GB)

The MacBook Air 13" M4 is a 24 GB laptop with more memory than speed: it runs 27–31B models at 5 tok/s. Among 24 GB machines, only the M5 Pro, M4 Pro and M5 write faster. macOS keeps about 3.5 GB, leaving 20.5 GB for models. Prompt reading is its weak spot: a 4K-token prompt waits 21 s before the first word on Gemma 4 26B A4B IT. At its $1,399 MSRP it gives 9.4 tok/s per $1,000, 7th of the 8 machines compared here. M4 generation, released 2025; the newer M5 writes 28% faster with 1.9× the compute.

Speed in 24 GB#4120 GB/s

Prompt reading#49 TFLOPS

For models20.5 GBmacOS keeps 3.5 GB

ChipM4released 2025

MacBook Air 13" M4 strengths and weaknesses

Strengths

Runs 35B MoE models: Qwen3.6 35B A3B at 25.7 tok/s in 19.8 GB

Weaknesses

70B models don't fit: Llama 3.3 70B Instruct needs 51.3 GB at Q4 vs 20.5 GB usable here
Slow on large dense models: Qwen3.8 27B at 4.8 tok/s
Slow prompt reading: A 32K-token prompt on Gemma 4 26B A4B IT waits 2.8 min for the first token
No CUDA: vLLM, AWQ, GPTQ and FP8 builds don't run; use GGUF or MLX

Memory for models

20.5 GB of 24 GB

9B6.6 GB14B11.9 GB27–31B17.9 GB35B MoE19.8 GB

9B6.6 GB14B11.9 GB35B MoE19.8 GB

For models 20.5 GBmacOS keeps 3.5 GBModel size at Q4

70B models need 51.3 GB at Q4, more than the 20.5 GB available. The 32 GB config fits Qwen3.8 27B →Or rent an H100 (80 GB) from $2.69/hr →

Speed per dollar and per watt

NewUsed

On Qwen3.5 9B (9B params) at Q4_K_M. New means a shop price, either a street quote or the maker's own price; used is compared with used. The last two columns are the speed you get per $1,000 of price and per 100 W of power; higher is better. The MacBook Air 13" M4 has no current new price, so it is shown at its MSRP.

Machine	New price	tok/s	tok/s per $1,000	tok/s per 100 W
Mac Mini M4 Pro24 GB	about $1,599	30.0	18.8	21.4
Mac Studio M4 Max36 GB	about $2,499	45.1	18.0	31.1
MacBook Pro 14" M4 Pro24 GB	about $1,999	30.0	15.0	–
MacBook Pro 14 M5 Pro64 GB	about $2,299	33.8	14.7	–
MacBook Air 13" M416 GB	about $999	13.2	13.2	–
MacBook Pro 16" M4 Pro24 GB	about $2,499	30.0	12.0	–
MacBook Air 13" M424 GB	about $1,399	13.2	9.4	–
MacBook Pro 14 M532 GB	about $1,839	16.9	9.2	–

Machine	Used price	tok/s	tok/s per $1,000	tok/s per 100 W
Mac Studio M4 Max36 GB	about $2,000	45.1	22.6	31.1
MacBook Pro 14 M5 Pro64 GB	about $2,109	33.8	16.0	–

Cost per token: the MacBook Air 13" M4 against 16–64 GB machines and GPUs

On Qwen3.5 9B (9B params, Q4_K_M, 4K context). Speed, cost per token and power per token depend on the model.

NewUsed

Speed$ per 1M tokens

≤16 GB24–36 GB48–64 GBthis laptopat MSRP

Cost per 1M output tokenson Qwen3.5 9B · card over 3 years + power

$6.72

Machine, MSRP: $6.72
Power: not published

3-year cost of ownership$1,399

Or rent an RTX 5090 in the cloud$0.46/hr

Renting costs as much as buying after ≈3,000 hours (≈2.1 years at 4 h a day).

Power$/kWhUseh/dayKeepyears

Each dot is a machine or GPU; hover or tap it for its numbers, click to open it. Machines without a price for the chosen basis are left out. Machines without a published power figure are left out of Per watt and costed without power. How we calculate cost per token

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, 8 GB less memory for $400 lessMacBook Air 13" M4 (16 GB)Apple · 12.5 GB for models · M4

same speed▼ $400 less▼ 8 GB$999

Same family Same chip, 8 GB more memoryMacBook Air 13" M4 (32 GB)Apple · 28.5 GB for models · M4

same speed▲ 8 GBno price

Same family 127% faster, same memoryMac Mini M4 Pro (24 GB)Apple · 20.5 GB for models · M4 Pro

▲ 127% speedsame memory$1,599

28% faster, 8 GB more memoryMacBook Pro 14 M5Apple · 28.5 GB for models · M5

▲ 28% speed▲ 8 GB$1,839

242% faster, 12 GB more memoryMac Studio M4 Max (36 GB)Apple · 32.5 GB for models · M4 Max

▲ 242% speed▲ 12 GB$2,499

Reads prompts 4× faster, 60.5 GB for modelsMacBook Pro 14 M5 ProApple · 60.5 GB for models · M5 Pro

▲ 4× prompts▲ 156% speed▲ 40 GB$2,299

MacBook Air 13" M4 tokens per second by model size

Output speed depends on memory bandwidth (120 GB/s). Grey rows need more than the 20.5 GB available.

9BQwen3.5 9B · Q4_K_M · 6.6 GB
13.2tok/s
14BQwen3 14B · Q4_K_M · 11.9 GB slow
8.8tok/s
27–31BQwen3.8 27B · Q4_K_M · 17.9 GB slow
4.8tok/s
35B MoEQwen3.6 35B A3B · IQ4_XS · 19.8 GB tight · IQ4_XS
25.7tok/s
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 16.9 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 1125 models on the MacBook Air 13" M4

Featured only

Best quant that fits in 20.5 GB, ordered by llmrun Score. Fit grades how well it runs here, S best. How ranking works

LLM models compatible with MacBook Air 13" M4 (24 GB) — ranked by Score, recency and Fit
Model	Quant	VRAM	Speed	Context	Fit
Gemma4 12B Mtp Assistant12BChat Q4_K_M·10.6 t/s	Q4_K_M	10.8 GB53%	10.6 t/s	—	B49, Decent
Gpt2137MChat Q4_K_M·933.3 t/s	Q4_K_M	0.1 GB0%	933.3 t/s	1K	S100, Runs great
DeepSeek R1 Distill Qwen 1.5B1.8BChatReasoningFunctions Q4_K_M·59.2 t/s	Q4_K_M	1.8 GB9%	59.2 t/s	131K	A83, Runs well
Hunyuan 0.5B Pretrain539MChat Q4_K_M·116.7 t/s	Q4_K_M	1.4 GB7%	116.7 t/s	262K	S94, Runs great
Jina Ocr V13.4BVision Q4_K_M·79.5 t/s	Q4_K_M	3.3 GB16%	79.5 t/s	33K	S88, Runs great
Qwen3 8B8.2BChatFunctions Q4_K_M·15.2 t/s	Q4_K_M	7.6 GB37%	15.2 t/s	41K	B55, Decent
Qwen3.5 4B Claude 4.6 Opus Reasoning Distilled4.7BChatReasoning Q4_K_M·25.8 t/s	Q4_K_M	3.4 GB17%	25.8 t/s	262K	A66, Runs well
GaMS3 12B Instruct11.8BChat Q4_K_M·10.4 t/s	Q4_K_M	8.9 GB43%	10.4 t/s	131K	B49, Decent
Granite Switch 4.1 3B Preview4.1BChat BF16·9.6 t/s	BF16	10.0 GB49%	9.6 t/s	131K	B47, Decent
HunyuanOCR1.1BVision Q4_K_M·78.5 t/s	Q4_K_M	1.8 GB9%	78.5 t/s	131K	S87, Runs great
Carbon 8B8.3BChat Q4_K_M·15.2 t/s	Q4_K_M	7.4 GB36%	15.2 t/s	33K	B55, Decent
Vllm Translategemma 12B IT13.2BChatTranslation Q4_K_M·9.4 t/s	Q4_K_M	9.8 GB48%	9.4 t/s	131K	B47, Decent
VibeThinker 1.5B1.8BChatMathCode Q4_K_M·59.2 t/s	Q4_K_M	1.8 GB9%	59.2 t/s	131K	A83, Runs well
Tiny GPT2LMHeadModel412008Chat Q4_K_M·8400.0 t/s	Q4_K_M	0.0 GB0%	8400.0 t/s	1K	S100, Runs great
Soren 1 Small1.9BChatReasoningCodeMath BF16·20.2 t/s	BF16	4.3 GB21%	20.2 t/s	1049K	B60, Decent

MacBook Air 13" M4 specs

Specs

Chip: M4
CPU cores: 10
GPU cores: 10
Unified memory: 24 GB
For models: 20.5 GB (macOS keeps 3.5 GB)
Memory bandwidth: 120 GB/s
FP16 (dense): 8.6 TFLOPS
Neural Engine: 38 TOPS
Released: 2025-03-12
MSRP: $1,399

MacBook Air 13" M4 FAQ

Is the MacBook Air 13" M4 (24 GB) worth it for local LLMs in 2026?

At its $1,399 MSRP it gives 9.4 tok/s per $1,000 on Qwen3.5 9B, 7th of the 8 machines compared here. It runs 27–31B models at 5 tok/s; 9B and 14B fit but are slow (13 and 9 tok/s), and 70B doesn't fit in 20.5 GB.

How much memory is available for AI on the MacBook Air 13" M4 (24 GB)?

About 20.5 GB of the 24 GB. macOS keeps roughly 3.5 GB for itself and other apps.

What is the largest model the MacBook Air 13" M4 (24 GB) can run?

Qwen3.6 35B A3B (36.0B) is the largest at 4-bit or better: at IQ4_XS it uses 19.8 GB, 97% of the available memory, and runs at about 25.7 tok/s.

Can the MacBook Air 13" M4 (24 GB) run Gemma 4 26B A4B IT?

Yes. Gemma 4 26B A4B IT fits at Q4_K_M, using 16.5 GB of the 20.5 GB available, and it is the best model that runs well on this machine.

Can the MacBook Air 13" M4 (24 GB) run 70B models?

No. Llama 3.3 70B Instruct needs 51.3 GB at Q4, more than the 20.5 GB available.

How long does the MacBook Air 13" M4 (24 GB) take to read a long prompt?

On Gemma 4 26B A4B IT, about 21 s for a 4K-token prompt and 2.8 min for a 32K-token document before the first word appears.

Is renting a GPU cheaper than buying the MacBook Air 13" M4?

An RTX 5090 (32 GB) rents from $0.46/hr. Renting costs as much as buying after ≈3,000 hours (≈2.1 years at 4 h a day).

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