Apple · M5 Ultra · Desktop

Best AI Models for Mac Studio M5 Ultra (512 GB)

Runs 70–120B models well; fits models up to 822B, like Muse Spark Open 816B A42B.

512 GB unified508.5 GB for models1.2 TB/s136 TFLOPS FP162026

Memory

96 GB$5,499 256 GB$9,499 512 GB$22,299

What LLMs can the Mac Studio M5 Ultra run?

#	Model	Speed	Quant · Memory
1	Inkling 2 SmallThinking MachinesHow to run → Q4_K_M · 359.1 GB · 46.5 t/s	46.5 tok/s	Q4_K_M · 359.1 GB
2	DeepSeek V4.1 Flash 1017DeepSeek Q4_K_M · 458.4 GB · 52.1 t/s	52.1 tok/s	Q4_K_M · 458.4 GB
3	MiMo V2.6 Flash RLXiaomi Q4_K_M · 186.5 GB · 49.8 t/s	49.8 tok/s	Q4_K_M · 186.5 GB
4	GLM 5.3Z.ai Q4_K_M · 453.8 GB · 22.2 t/s	22.2 tok/s	Q4_K_M · 453.8 GB
5	Hy3Tencent Q4_K_M · 182.3 GB · 32.6 t/s	32.6 tok/s	Q4_K_M · 182.3 GB

#	Model	Speed	Quant · Memory
1	DeepSeek V4.1 Flash 1017DeepSeekHow to run → Q4_K_M · 458.4 GB · 52.1 t/s	52.1 tok/s	Q4_K_M · 458.4 GB
2	GLM 5.3Z.ai Q4_K_M · 453.8 GB · 22.2 t/s	22.2 tok/s	Q4_K_M · 453.8 GB
3	Muse Spark Open 816B A42BMeta Q4_0 · 461.5 GB · 27.4 t/s	27.4 tok/s	Q4_0 · 461.5 GB
4	MiMo V2.6 Flash RLXiaomi Q4_K_M · 186.5 GB · 49.8 t/s	49.8 tok/s	Q4_K_M · 186.5 GB
5	Hy3Tencent Q4_K_M · 182.3 GB · 32.6 t/s	32.6 tok/s	Q4_K_M · 182.3 GB

#	Model	Speed	Quant · Memory
1	DeepSeek V4.1 Flash 1017DeepSeekHow to run → Q4_K_M · 458.4 GB · 52.1 t/s	52.1 tok/s	Q4_K_M · 458.4 GB
2	Inkling 2 SmallThinking Machines Q4_K_M · 359.1 GB · 46.5 t/s	46.5 tok/s	Q4_K_M · 359.1 GB
3	DeepSeek V4.1 FlashDeepSeek Q4_K_M · 458.4 GB · 52.1 t/s	52.1 tok/s	Q4_K_M · 458.4 GB
4	GLM 5.3Z.ai Q4_K_M · 453.8 GB · 22.2 t/s	22.2 tok/s	Q4_K_M · 453.8 GB
5	Hy3Tencent Q4_K_M · 182.3 GB · 32.6 t/s	32.6 tok/s	Q4_K_M · 182.3 GB

#	Model	Speed	Quant · Memory
1	DeepSeek V4.1 Flash 1017DeepSeekHow to run → Q4_K_M · 458.4 GB · 52.1 t/s	52.1 tok/s	Q4_K_M · 458.4 GB
2	DeepSeek V4.1 FlashDeepSeek Q4_K_M · 458.4 GB · 52.1 t/s	52.1 tok/s	Q4_K_M · 458.4 GB
3	GLM 5.3Z.ai Q4_K_M · 453.8 GB · 22.2 t/s	22.2 tok/s	Q4_K_M · 453.8 GB
4	Muse Spark Open 816B A42BMeta Q4_0 · 461.5 GB · 27.4 t/s	27.4 tok/s	Q4_0 · 461.5 GB
5	Hy3Tencent Q4_K_M · 182.3 GB · 32.6 t/s	32.6 tok/s	Q4_K_M · 182.3 GB

#	Model	Speed	Quant · Memory
1	Inkling 2 SmallThinking MachinesHow to run → Q4_K_M · 359.1 GB · 46.5 t/s	46.5 tok/s	Q4_K_M · 359.1 GB
2	MiMo V2.6 Flash RLXiaomi Q4_K_M · 186.5 GB · 49.8 t/s	49.8 tok/s	Q4_K_M · 186.5 GB
3	DeepSeek V4 Flash 0731DeepSeek Q4_K_M · 183.0 GB · 36.1 t/s	36.1 tok/s	Q4_K_M · 183.0 GB
4	GLM 5.3 FlashZ.ai Q4_K_M · 205.2 GB · 45.6 t/s	45.6 tok/s	Q4_K_M · 205.2 GB
5	Hy3Tencent Q4_K_M · 182.3 GB · 32.6 t/s	32.6 tok/s	Q4_K_M · 182.3 GB

#	Model	Speed ↓	Quant · Memory
1	Qwen3.5 4BAlibabaHow to run → Q4_K_M · 3.4 GB · 264 t/s	264 tok/s	Q4_K_M · 3.4 GB
2	Qwen3.5 9BAlibaba Q4_K_M · 6.6 GB · 135 t/s	135 tok/s	Q4_K_M · 6.6 GB
3	GPT OSS 20BOpenAI Q4_K_M · 13.1 GB · 128 t/s	128 tok/s	Q4_K_M · 13.1 GB
4	Gemma 4 26B A4B ITGoogle Q4_K_M · 16.5 GB · 109 t/s	109 tok/s	Q4_K_M · 16.5 GB
5	Hy3Tencent Q4_K_M · 182.3 GB · 32.6 t/s	32.6 tok/s	Q4_K_M · 182.3 GB

#	Model	Speed	Quant · Memory
1	Inkling 2 SmallThinking MachinesHow to run → Q4_K_M · 359.1 GB · 46.5 t/s	46.5 tok/s	Q4_K_M · 359.1 GB
2	DeepSeek V4.1 Flash 1017DeepSeek Q4_K_M · 458.4 GB · 52.1 t/s	52.1 tok/s	Q4_K_M · 458.4 GB
3	DeepSeek V4.1 FlashDeepSeek Q4_K_M · 458.4 GB · 52.1 t/s	52.1 tok/s	Q4_K_M · 458.4 GB
4	Muse Spark Open 816B A42BMeta Q4_0 · 461.5 GB · 27.4 t/s	27.4 tok/s	Q4_0 · 461.5 GB
5	Muse Glimmer 30BMeta Q4_K_M · 19.6 GB · 46.9 t/s	46.9 tok/s	Q4_K_M · 19.6 GB

#	Model	Speed	Quant · Memory
1	DeepSeek V4.1 Flash 1017DeepSeekHow to run → Q4_K_M · 458.4 GB · 52.1 t/s	52.1 tok/s	Q4_K_M · 458.4 GB
2	GLM 5.3Z.ai Q4_K_M · 453.8 GB · 22.2 t/s	22.2 tok/s	Q4_K_M · 453.8 GB
3	GLM 5.2Z.ai Q4_K_M · 453.8 GB · 22.2 t/s	22.2 tok/s	Q4_K_M · 453.8 GB
4	DeepSeek V4.1 FlashDeepSeek Q4_K_M · 458.4 GB · 52.1 t/s	52.1 tok/s	Q4_K_M · 458.4 GB
5	Hy3Tencent Q4_K_M · 182.3 GB · 32.6 t/s	32.6 tok/s	Q4_K_M · 182.3 GB

See all 443 models on the Mac Studio M5 Ultra →Bars = traction for models without a Score yet

About the Mac Studio M5 Ultra (512 GB)

The Mac Studio M5 Ultra is the fastest 512 GB machine: it runs 70–120B models well. macOS keeps about 3.5 GB, leaving 508.5 GB for models. With 136 TFLOPS it reads long prompts faster than most 512 GB machines: a 4K-token prompt takes 5.1 s. At its $22,299 MSRP it gives 3.9 tok/s per $1,000, 6th of the 6 machines compared here. M5 Ultra generation, released 2026.

Speed in 512 GB#11.2 TB/s

Prompt reading#1136 TFLOPS

For models508.5 GBmacOS keeps 3.5 GB

ChipM5 Ultrareleased 2026

Mac Studio M5 Ultra strengths and weaknesses

Strengths

Runs 120B MoE models: GPT OSS 120B at 89.0 tok/s in 70.8 GB
Room for long context: Llama 3.3 70B Instruct at Q4 uses 51.3 GB, leaving 457.2 GB free

Weaknesses

Slow prompt reading: A 32K-token prompt on Inkling 2 Small waits 41.0 s for the first token
No CUDA: vLLM, AWQ, GPTQ and FP8 builds don't run; use GGUF or MLX

Memory for models

508.5 GB of 512 GB

120B MoE70.8 GB822B461.5 GB

For models 508.5 GBmacOS keeps 3.5 GBModel size at Q4

Everything on the size ladder fits. MiMo V2.6 Flash RL and DeepSeek V4 Flash 0731 together use 369.5 GB and leave 139 GB free.

Speed per dollar and per watt

NewUsed

On Qwen3.6 35B A3B (35B params, 3B active) 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 Mac Studio M5 Ultra 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
RTX 509032 GB	about $5,502	213	38.8	37.1
Mac Studio M5 Ultra96 GB	about $5,499	87.9	16.0	–
Mac Studio M2 Ultra192 GB	about $5,599	75.9	13.6	25.7
Mac Studio M5 Ultra256 GB	about $9,499	87.9	9.3	–
Mac Studio M3 Ultra512 GB	about $9,499	76.6	8.1	28.4
Mac Studio M5 Ultra512 GB	about $22,299	87.9	3.9	–

Machine	Used price	tok/s	tok/s per $1,000	tok/s per 100 W
RTX 409024 GB	about $3,000	151	50.3	33.6
RTX 509032 GB	about $7,000	213	30.5	37.1

Cost per token: the Mac Studio M5 Ultra against 24–512 GB machines and GPUs

On Qwen3.6 35B A3B (35B params, 3B active, Q4_K_M, 4K context). Speed, cost per token and power per token depend on the model.

NewUsed

Speed$ per 1M tokens

24–36 GB96–128 GB192+ GBthis desktopat MSRP

Cost per 1M output tokenson Qwen3.6 35B A3B · card over 3 years + power

$16.09

Machine, MSRP: $16.09
Power: not published

3-year cost of ownership$22,299

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 Qwen3.6 35B A3B.

Same chip, 256 GB less memory for $12,800 lessMac Studio M5 Ultra (256 GB)Apple · 252.5 GB for models · M5 Ultra

same speed▼ $12,800 less▼ 256 GB$9,499

Same family Same chip, 416 GB less memory for $16,800 lessMac Studio M5 Ultra (96 GB)Apple · 92.5 GB for models · M5 Ultra

same speed▼ $16,800 less▼ 416 GB$5,499

Same family Machine, 13% slowerMac Studio M3 Ultra (512 GB)Apple · 508.5 GB for models · M3 Ultra

▼ 13% speedsame memory$9,499

Step down to 188.5 GBMac Studio M2 Ultra (192 GB)Apple · 188.5 GB for models · M2 Ultra

▼ 14% speed▼ 320 GB$5,599

1.7× faster if your models fit in 24 GBRTX 4090NVIDIA · 24 GB VRAM · GPU

▲ 72% speed▼ 484.5 GB$1,599rent $0.34/hr

2.4× faster if your models fit in 32 GBRTX 5090NVIDIA · 32 GB VRAM · GPU

▲ 143% speed▼ 476.5 GB$5,502rent $0.50/hr

Mac Studio M5 Ultra tokens per second by model size

Output speed depends on memory bandwidth (1228 GB/s). Grey rows need more than the 508.5 GB available.

9BQwen3.5 9B · Q4_K_M · 6.6 GB
135tok/s
14BQwen3 14B · Q4_K_M · 11.9 GB
90.5tok/s
27–31BQwen3.8 27B · Q4_K_M · 17.9 GB
49.3tok/s
35B MoEQwen3.6 35B A3B · Q4_K_M · 22.0 GB
87.9tok/s
70BLlama 3.3 70B Instruct · Q4_K_M · 51.3 GB
18.5tok/s
120B MoEGPT OSS 120B · Q4_K_M · 70.8 GB
89.0tok/s

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

Ultra generations: writing vs compute

M1 Ultra → M5 Ultra

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

Writing speed follows memory bandwidth and barely moved: 75.9 to 87.9 tok/s since the M1 Ultra. Compute, which sets how fast prompts are read, grew 4.4×. Indexed to the M1 Ultra; speeds on Qwen3.6 35B A3B, same 128 GB.

All 443 models on the Mac Studio M5 Ultra

Featured only

Best quant that fits in 508.5 GB, ordered by llmrun Score. Fit grades how well it runs here, S best. How ranking works

LLM models compatible with Mac Studio M5 Ultra (512 GB) — ranked by Score, recency and Fit
Model	Quant	VRAM	Speed	Context	Fit
Qwen3 VL 4B Instruct4.4BVisionFunctions Q4_K_M·272.9 t/s	Q4_K_M	4.5 GB1%	272.9 t/s	262K	S100, Runs great
SmolVLM2 2.2B Instruct2.2BVision Q4_K_M·580.8 t/s	Q4_K_M	2.3 GB0%	580.8 t/s	8K	S100, Runs great
InternVL3 8B7.9BVision Q4_K_M·164.0 t/s	Q4_K_M	7.8 GB2%	164.0 t/s	—	S97, Runs great
Kimi VL A3B Instruct16.4BVisionFunctions Q4_K_M·126.2 t/s	Q4_K_M	10.7 GB2%	126.2 t/s	131K	S95, Runs great
OvisOCR3853MVisionFunctions Q4_K_M·999.5 t/s	Q4_K_M	0.9 GB0%	999.5 t/s	262K	S100, Runs great
Surya Ocr 2686MVision Q4_K_M·1131.1 t/s	Q4_K_M	0.8 GB0%	1131.1 t/s	262K	S100, Runs great
GLM 4.5 Air110.5BChatFunctions Q4_K_M·50.7 t/s	Q4_K_M	67.6 GB13%	50.7 t/s	131K	A80, Runs well
Llama 3.2 3B Instruct3.2BChatFunctions Q4_K_M·405.5 t/s	Q4_K_M	4.1 GB1%	405.5 t/s	131K	S100, Runs great
Moondream21.9BVision Q4_K_M·676.9 t/s	Q4_K_M	3.0 GB1%	676.9 t/s	—	S100, Runs great
SmolVLM2 256M Video Instruct256MVision Q4_K_M·1719.2 t/s	Q4_K_M	0.6 GB0%	1719.2 t/s	8K	S100, Runs great
MiniCPM OCR 1.01.3BVisionFunctions Q4_K_M·760.7 t/s	Q4_K_M	1.5 GB0%	760.7 t/s	262K	S100, Runs great
Falcon H1 7B Instruct7.6BChatFunctions Q4_K_M·170.6 t/s	Q4_K_M	6.3 GB1%	170.6 t/s	262K	S98, Runs great
Surya Ocr 2.5689MVision Q4_K_M·1131.1 t/s	Q4_K_M	0.8 GB0%	1131.1 t/s	262K	S100, Runs great
GLM 4.6356.8BChatFunctions Q4_K_M·23.7 t/s	Q4_K_M	216.9 GB43%	23.7 t/s	203K	B64, Decent
Moondream3 Preview9.3BVision BF16·42.2 t/s	BF16	23.1 GB5%	42.2 t/s	—	A76, Runs well

Mac Studio M5 Ultra specs

Specs

Chip: M5 Ultra
CPU cores: 36
GPU cores: 80
Unified memory: 512 GB
For models: 508.5 GB (macOS keeps 3.5 GB)
Memory bandwidth: 1.2 TB/s
FP16 (dense): 136 TFLOPS
Neural Engine: 76 TOPS
Released: 2026-10-28
MSRP: $22,299

Mac Studio M5 Ultra FAQ

Is the Mac Studio M5 Ultra (512 GB) worth it for local LLMs in 2026?

At its $22,299 MSRP it gives 3.9 tok/s per $1,000 on Qwen3.6 35B A3B, 6th of the 6 machines compared here. It runs 70–120B models well; fits models up to 822B, like Muse Spark Open 816B A42B.

How much memory is available for AI on the Mac Studio M5 Ultra (512 GB)?

About 508.5 GB of the 512 GB. macOS keeps roughly 3.5 GB for itself and other apps.

What is the largest model the Mac Studio M5 Ultra (512 GB) can run?

Muse Spark Open 816B A42B (822.0B) is the largest at 4-bit or better: at Q4_0 it uses 461.5 GB, 91% of the available memory, and runs at about 27.4 tok/s. At 3-bit, a bigger model fits too: Inkling (952.4B) at Q3_K_M, 467.9 GB.

Can the Mac Studio M5 Ultra (512 GB) run Inkling 2 Small?

Yes. Inkling 2 Small fits at Q4_K_M, using 359.1 GB of the 508.5 GB available, and it is the best model that runs well on this machine.

How long does the Mac Studio M5 Ultra (512 GB) take to read a long prompt?

On Inkling 2 Small, about 5.1 s for a 4K-token prompt and 41 s for a 32K-token document before the first word appears.

48 GB GPUs All Macs Desktops & mini PCs Best coding LLMs All hardware