128GB VRAM

Mac M4 Max 128G — what LLMs can it run?

75 of 81 indexed models fit comfortably in 128GB at 4K context, each at the highest-quality quant that still leaves headroom.

70B+ · 9

Mac M4 Max 128G — what LLMs can it run? — 70B+
Model	Quant	Est. VRAM	Headroom	tok/s
DBRX Instruct132B	Q4_K_M	84.31 GB	+43.7 GB	15
GPT-OSS 120B117B MoE	MXFP4	65.15 GB	+62.9 GB	22
Llama 4 Scout 17B (16E)109B MoE	Q4_K_M	69.88 GB	+58.1 GB	22
GLM-4.5-Air106B MoE	Q4_K_M	67.94 GB	+60.1 GB	18
Llama 3.2 90B Vision Instruct90B	Q4_K_M	57.5 GB	+70.5 GB	22
Qwen2.5 72B Instruct72B	Q5_K_M	55.31 GB	+72.7 GB	24
Llama 3.1 70B Instruct70B	Q5_K_M	53.75 GB	+74.3 GB	32
Llama 3.3 70B Instruct70B	Q5_K_M	53.75 GB	+74.3 GB	32
DeepSeek-R1-Distill-Llama-70B70B	Q4_K_M	46.1 GB	+81.9 GB	36

32B · 19

Mac M4 Max 128G — what LLMs can it run? — 32B
Model	Quant	Est. VRAM	Headroom	tok/s
Mixtral 8x7B Instruct47B MoE	Q4_K_M	30.13 GB	+97.9 GB	48
Seed-OSS 36B Instruct36B	Q5_K_M	27.81 GB	+100.2 GB	37
Command R 35B35B	Q4_K_M	22.86 GB	+105.1 GB	42
Yi 1.5 34B Chat34B	Q4_K_M	22.82 GB	+105.2 GB	40
Qwen3 32B Instruct32B	Q4_K_M	21.85 GB	+106.2 GB	42
Qwen2.5 32B Instruct32B	Q4_K_M	21.69 GB	+106.3 GB	44
Qwen2.5-Coder 32B Instruct32B	Q4_K_M	21.69 GB	+106.3 GB	44
DeepSeek-R1-Distill-Qwen-32B32B	Q4_K_M	21.69 GB	+106.3 GB	42
Qwen3 30B-A3B Instruct30B-A3B	Q5_K_M	23.04 GB	+105 GB	82
Qwen3-Coder 30B-A3B Instruct30B-A3B	Q5_K_M	23.04 GB	+105 GB	80
Qwen3-VL 30B-A3B Instruct30B-A3B	Q8_0	34.28 GB	+93.7 GB	78
Qwen3.8 27B27B	Q4_K_M	17.47 GB	+110.5 GB	—
Gemma 3 27B IT27B	Q4_K_M	19.49 GB	+108.5 GB	48
Gemma 2 27B Instruct27B	Q5_K_M	21.76 GB	+106.2 GB	42
Mistral Small 24B Instruct24B	EXL2 4.65bpw	15.26 GB	+112.7 GB	88
Devstral Small 1.1 24B24B	Q6_K	20.91 GB	+107.1 GB	48
Magistral Small 1.2 24B24B	Q6_K	20.91 GB	+107.1 GB	47
Codestral 22B22B	Q4_K_M	15.03 GB	+113 GB	58
GPT-OSS 20B21B MoE	MXFP4	11.81 GB	+116.2 GB	195

14B · 17

Mac M4 Max 128G — what LLMs can it run? — 14B
Model	Quant	Est. VRAM	Headroom	tok/s
InternLM2 20B Chat20B	Q5_K_M	15.59 GB	+112.4 GB	68
DeepSeek-Coder-V2-Lite Instruct16B	Q8_0	18.36 GB	+109.6 GB	118
DeepSeek-V2-Lite Chat16B	Q4_K_M	10.87 GB	+117.1 GB	142
StarCoder2 15B15B	Q4_K_M	10.19 GB	+117.8 GB	92
Qwen3 14B Instruct14B	Q5_K_M	11.65 GB	+116.4 GB	78
Qwen2.5 14B Instruct14B	Q5_K_M	11.73 GB	+116.3 GB	86
DeepSeek-R1-Distill-Qwen-14B14B	EXL2 4.65bpw	9.75 GB	+118.3 GB	128
Phi-4 14B14B	Q5_K_M	11.77 GB	+116.2 GB	78
Phi-3 Medium 14B Instruct14B	Q6_K	12.86 GB	+115.1 GB	88
Mistral Nemo 12B Instruct12B	Q6_K	11.14 GB	+116.9 GB	95
Gemma 3 12B IT12B	Q5_K_M	10.6 GB	+117.4 GB	92
Stable LM 2 12B Chat12B	Q4_K_M	8.35 GB	+119.7 GB	108
Jamba 1.5 Mini12B	Q4_K_M	8.15 GB	+119.9 GB	95
Llama 3.2 11B Vision Instruct11B	Q8_0	12.9 GB	+115.1 GB	72
Solar 10.7B Instruct11B	Q4_K_M	7.6 GB	+120.4 GB	125
Falcon 3 10B Instruct10B	Q4_K_M	7.28 GB	+120.7 GB	118
Gemma 2 9B Instruct9B	Q8_0	11.7 GB	+116.3 GB	108

7B · 20

Mac M4 Max 128G — what LLMs can it run? — 7B
Model	Quant	Est. VRAM	Headroom	tok/s
GLM-4-9B-Chat9B	Q8_0	10.16 GB	+117.8 GB	105
Qwen3-VL 8B Instruct8B	Q8_0	10.39 GB	+117.6 GB	108
Ministral 3 8B Instruct8B	Q8_0	10.02 GB	+118 GB	—
Qwen2-VL 7B Instruct7B	Q4_K_M	5.49 GB	+122.5 GB	72
Granite 3.1 8B Instruct8B	Q4_K_M	5.88 GB	+122.1 GB	142
Qwen3 8B Instruct8B	Q6_K	7.64 GB	+120.4 GB	122
Llama 3.1 8B Instruct8B	Q8_0	9.47 GB	+118.5 GB	118
Nous Hermes 3 Llama 3.1 8B8B	EXL2 4.65bpw	5.43 GB	+122.6 GB	232
Aya 23 8B8B	Q4_K_M	5.64 GB	+122.4 GB	145
OpenChat 3.6 8B8B	EXL2 4.65bpw	5.43 GB	+122.6 GB	228
DeepSeek-R1-Distill-Llama-8B8B	Q5_K_M	6.51 GB	+121.5 GB	128
InternLM2 7B Chat7B	Q4_K_M	5.45 GB	+122.6 GB	148
Qwen2.5 7B Instruct7B	Q6_K	6.77 GB	+121.2 GB	132
Qwen2.5-Coder 7B Instruct7B	EXL2 4.65bpw	4.87 GB	+123.1 GB	248
WizardLM-2 7B7B	Q4_K_M	5.07 GB	+122.9 GB	152
DeepSeek-R1-Distill-Qwen-7B7B	EXL2 4.65bpw	4.87 GB	+123.1 GB	210
OLMo 2 7B Instruct7B	Q8_0	10.3 GB	+117.7 GB	125
Mistral 7B Instruct v0.37B	Q6_K	6.75 GB	+121.3 GB	135
Zephyr 7B Beta7B	Q6_K	6.75 GB	+121.3 GB	132
Gemma 3 4B IT4B	Q8_0	5.36 GB	+122.6 GB	145

≤3B · 10

Mac M4 Max 128G — what LLMs can it run? — ≤3B
Model	Quant	Est. VRAM	Headroom	tok/s
Qwen3 4B Instruct4B	Q6_K	4.05 GB	+124 GB	145
Phi-4 Mini Instruct3.8B	Q8_0	4.68 GB	+123.3 GB	262
Phi-3.5 Mini Instruct3.8B	Q8_0	5.89 GB	+122.1 GB	255
Llama 3.2 3B Instruct3B	Q8_0	4.05 GB	+124 GB	285
Qwen2.5 3B Instruct3B	Q8_0	3.67 GB	+124.3 GB	290
Gemma 2 2B Instruct2B	Q8_0	3.34 GB	+124.7 GB	320
Qwen3 1.7B Instruct1.7B	Q8_0	2.39 GB	+125.6 GB	240
Qwen2.5 1.5B Instruct1.5B	Q8_0	1.83 GB	+126.2 GB	410
Llama 3.2 1B Instruct1B	Q8_0	1.51 GB	+126.5 GB	450
Qwen2.5 0.5B Instruct0.5B	Q8_0	0.6 GB	+127.4 GB	540

How this list is built

Each row is the lowest-perplexity-loss quant of that model whose estimated total — weights plus KV cache at 4K context plus activation buffer — uses at most 88% of the card. That is the calculator's "green" threshold, so every row here has real headroom rather than only just fitting. Raise the context length and the list shortens; the calculator lets you check any combination directly.

A further 1 models load but with no headroom to spare (up to 105% of VRAM) — the Quant Hub’s GPU chips count those too, which is why its number is higher.

Open this card in the VRAM calculator Browse all 81 models

Measured on this card

No benchmark runs in this index were recorded on a Mac M4 Max 128G. Every figure on this page is calculated from the model architecture and the quant level — treat them as estimates, not measurements.

Cards with the same budget

What fits is decided by memory, so every 128GB card of this type returns the same list. These pages are not different answers — they differ in throughput, which this index does not measure per card.

Stepping up

A Mac M3 Ultra 192G (192GB) fits 1 more of the indexed models than this card. Mac M3 Ultra 192G →