llmrun

Apple · M5 Max · Desktop

Best AI Models for Mac Studio M5 Max (128 GB)

Runs 27–31B and 120B MoE models well; dense 70B fits but is slow at 9 tok/s.

128 GB unified124.5 GB for models614 GB/s68 TFLOPS FP162026
Memory
36 GB48 GB64 GB128 GB
Same chip in
Mac StudioMacBook Pro 16"about $5,399MacBook Pro 14"

What LLMs can the Mac Studio M5 Max run?

#ModelSpeedllmrun ScoreQuant · Memory
1
Qwen3.8 27BAlibabaHow to run →
Q4_K_M · 17.9 GB
24.7 tok/sQ4_K_M · 17.9 GB
2
Qwen3.6 27BAlibaba
Q4_K_M · 17.9 GB
24.7 tok/sQ4_K_M · 17.9 GB
3
Gemma 4 31B ITGoogle
Q4_K_M · 21.7 GB
21.1 tok/sQ4_K_M · 21.7 GB
4
Gemma 4 26B A4B ITGoogle
Q4_K_M · 16.5 GB
80.7 tok/sQ4_K_M · 16.5 GB
5
NVIDIA Nemotron 3.5 Lightning 30B A3B BF16NVIDIA
Q4_K_M · 19.3 GB
72.7 tok/sQ4_K_M · 19.3 GB
See all 367 models on the Mac Studio M5 Max →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 Mac Studio M5 Max (128 GB)

The Mac Studio M5 Max is one of the fastest 128 GB machines: it runs 27–31B and 120B MoE models well. Among 128 GB machines, only the M1 Ultra and M2 Ultra write faster. macOS keeps about 3.5 GB, leaving 124.5 GB for models. With 68 TFLOPS it reads long prompts faster than most 128 GB machines: a 4K-token prompt takes 17 s. M5 Max generation, released 2026.

Speed in 128 GB#3614 GB/s
Prompt reading#368 TFLOPS
For models124.5 GBmacOS keeps 3.5 GB
ChipM5 Maxreleased 2026

Mac Studio M5 Max strengths and weaknesses

Strengths

  • Runs 120B MoE models: GPT OSS 120B at 65.5 tok/s in 70.8 GB
  • Room for long context: Qwen3.8 27B at Q4 uses 17.9 GB, leaving 106.6 GB free

Weaknesses

  • Slow on large dense models: Llama 3.3 70B Instruct at 9.2 tok/s
  • Slow prompt reading: A 32K-token prompt on Qwen3.8 27B waits 2.2 min for the first token
  • No CUDA: vLLM, AWQ, GPTQ and FP8 builds don't run; use GGUF or MLX

Memory for models

124.5 GB of 128 GB
35B MoE22 GB70B51.3 GB123B74.7 GB
35B MoE22 GB123B74.7 GB
For models 124.5 GBmacOS keeps 3.5 GBModel size at Q4

Everything on the size ladder fits. Two 27B-class models together use 35.7 GB and leave 88.8 GB free.

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, 64 GB less memoryMac Studio M5 Max (64 GB)Apple · 60.5 GB for models · M5 Max
same speed▼ 64 GBno price
Same familySame chip, 80 GB less memoryMac Studio M5 Max (48 GB)Apple · 44.5 GB for models · M5 Max
same speed▼ 80 GBno price
Same familyReads prompts 4× faster, 120 GB for modelsDGX SparkNVIDIA · 120 GB for models · GB10
▲ 4× prompts▼ 20% speed▼ 4.5 GBabout $3,999
12% faster, 64 GB more memoryMac Studio M2 Ultra (192 GB)Apple · 188.5 GB for models · M2 Ultra
▲ 12% speed▲ 64 GBabout $5,599
2.2× faster if your models fit in 24 GBRTX 4090NVIDIA · 24 GB VRAM · GPU
▲ 123% speed▼ 100.5 GBabout $1,599rent $0.34/hr
3.1× faster if your models fit in 32 GBRTX 5090NVIDIA · 32 GB VRAM · GPU
▲ 215% speed▼ 92.5 GBabout $5,376rent $0.59/hr

Mac Studio M5 Max tokens per second by model size

Output speed depends on memory bandwidth (614 GB/s). Grey rows need more than the 124.5 GB available.

  1. 9BQwen3.5 9B · Q4_K_M · 6.6 GB
    67.6tok/s
  2. 14BQwen3 14B · Q4_K_M · 11.9 GB
    45.2tok/s
  3. 27–31BQwen3.8 27B · Q4_K_M · 17.9 GB
    24.7tok/s
  4. 35B MoEQwen3.6 35B A3B · Q4_K_M · 22.0 GB
    67.8tok/s
  5. 70BLlama 3.3 70B Instruct · Q4_K_M · 51.3 GB slow
    9.2tok/s
  6. 120B MoEGPT OSS 120B · Q4_K_M · 70.8 GB
    65.5tok/s

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

Max generations: writing vs compute

M1 Max → M5 Max
Writing speed (tok/s)Compute (TFLOPS FP16)
1×2×3×4×5×M1 Max2022M2 Max2023M3 Max2023M4 Max2025M5 Max2026M1 Max · 15.6 TFLOPS FP16M1 Max · 54.5 tok/s · 400 GB/sM2 Max · 27.2 TFLOPS FP16M2 Max · 54.5 tok/s · 400 GB/sM3 Max · 28.7 TFLOPS FP16M3 Max · 55.2 tok/s · 409.6 GB/sM4 Max · 34.4 TFLOPS FP16M4 Max · 64.2 tok/s · 546 GB/sM5 Max · 68 TFLOPS FP16M5 Max · 67.8 tok/s · 614 GB/sCompute 4.4×Writing +24%
1×2×3×4×5×M12022M22023M32023M42025M52026M1 Max · 15.6 TFLOPS FP16M1 Max · 54.5 tok/s · 400 GB/sM2 Max · 27.2 TFLOPS FP16M2 Max · 54.5 tok/s · 400 GB/sM3 Max · 28.7 TFLOPS FP16M3 Max · 55.2 tok/s · 409.6 GB/sM4 Max · 34.4 TFLOPS FP16M4 Max · 64.2 tok/s · 546 GB/sM5 Max · 68 TFLOPS FP16M5 Max · 67.8 tok/s · 614 GB/sCompute 4.4×Writing +24%

Writing speed follows memory bandwidth and barely moved: 54.5 to 67.8 tok/s since the M1 Max. Compute, which sets how fast prompts are read, grew 4.4×. Indexed to the M1 Max; speeds on Qwen3.6 35B A3B, same 64 GB.

All 1228 models on the Mac Studio M5 Max

Featured only

Best quant that fits in 124.5 GB, ordered by llmrun Score. Fit grades how well it runs here, S best. How ranking works

LLM models compatible with Mac Studio M5 Max (128 GB) — ranked by Score, recency and Fit
ModelVRAMScoreFit
Qwen2 VL 72B Instruct
73.4B·Q4_K_M·9.5 t/s·33K ctx
49.7 GBB47, Decent
RedPajama INCITE 7B Base
7B·Q4_K_M·93.0 t/s·2K ctx
4.6 GBS90, Runs great
Xgen 7B 8k Base
7B·Q4_K_M·93.0 t/s·8K ctx
5.7 GBS90, Runs great
Tiny Random LlamaForCausalLM
1M·Q4_K_M·42980.0 t/s
0.2 GBS100, Runs great
MagicPrompt Stable Diffusion
137M·Q4_K_M·4775.6 t/s·1K ctx
0.1 GBS100, Runs great
Meta Llama 3 8B Instruct
8.0B·Q4_K_M·79.7 t/s·8K ctx
6.2 GBS88, Runs great
Gemma 2B IT
2.5B·Q4_K_M·260.5 t/s
3.5 GBS100, Runs great
Llama Guard 3 8B
8.0B·BF16·24.3 t/s
20.2 GBA65, Runs well
OpenReasoning Nemotron 32B
32.8B·Q4_K_M·21.0 t/s·131K ctx
24.3 GBB61, Decent
Gemma 2 2B
2.6B·Q4_K_M·248.4 t/s
3.6 GBS100, Runs great
Qwen 72B
72.3B·Q4_K_M·9.0 t/s·33K ctx
52.5 GBB47, Decent
Qwen2.5 Coder 3B
3.1B·Q4_K_M·192.7 t/s·33K ctx
2.8 GBS99, Runs great
Nemotron Labs Diffusion 8B Base
8.5B·BF16·24.5 t/s·4K ctx
17.9 GBA65, Runs well
Qwen2 0.5B
494M·Q4_K_M·693.2 t/s·131K ctx
0.8 GBS100, Runs great
Nemotron H 8B Reasoning 128K
8.1B·BF16·24.1 t/s
20.4 GBB64, Decent

Mac Studio M5 Max specs

Specs

Chip
M5 Max
CPU cores
18
GPU cores
40
Unified memory
128 GB
For models
124.5 GB (macOS keeps 3.5 GB)
Memory bandwidth
614 GB/s
FP16 (dense)
68 TFLOPS
Neural Engine
38 TOPS
Released
2026-09-22

Mac Studio M5 Max FAQ

How much memory is available for AI on the Mac Studio M5 Max (128 GB)?

About 124.5 GB of the 128 GB. macOS keeps roughly 3.5 GB for itself and other apps.

What is the largest model the Mac Studio M5 Max (128 GB) can run?

NVIDIA Nemotron 3 Super 120B A12B BF16 (123.6B) at Q4_K_M uses 74.7 GB, 60% of the available memory, and runs at about 27.9 tok/s.

Can the Mac Studio M5 Max (128 GB) run Qwen3.8 27B?

Yes. Qwen3.8 27B fits at Q4_K_M, using 17.9 GB of the 124.5 GB available, and it is the best model that runs well on this machine.

Can the Mac Studio M5 Max (128 GB) run 70B models?

Yes, but slowly: Llama 3.3 70B Instruct at Q4_K_M uses 51.3 GB and writes about 9.2 tok/s.

How long does the Mac Studio M5 Max (128 GB) take to read a long prompt?

On Qwen3.8 27B, about 17 s for a 4K-token prompt and 2.2 min for a 32K-token document before the first word appears.

Is renting a GPU cheaper than buying the Mac Studio M5 Max?

An RTX PRO 6000 Blackwell Max-Q Workstation Edition (96 GB) rents from $1.22/hr.

48 GB GPUsAll MacsDesktops & mini PCsMac Studio M3 UltraMac Pro M2 UltraMac Studio M1 UltraMac Studio M2 UltraBest coding LLMsAll hardware