llmrun

Apple · M5 Max · Desktop

Best AI Models for Mac Studio M5 Max (36 GB)

Runs 27–35B models well; 70B doesn't fit in 32.5 GB.

36 GB unified32.5 GB for models460 GB/s54 TFLOPS FP162026
Memory
36 GB48 GB64 GB128 GB
Same chip in
Mac StudioMacBook Pro 16"MacBook 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
18.5 tok/sQ4_K_M · 17.9 GB
2
Qwen3.6 27BAlibaba
Q4_K_M · 17.9 GB
18.5 tok/sQ4_K_M · 17.9 GB
3
Gemma 4 31B ITGoogle
Q4_K_M · 21.7 GB
15.8 tok/sQ4_K_M · 21.7 GB
4
Gemma 4 26B A4B ITGoogle
Q4_K_M · 16.5 GB
68.8 tok/sQ4_K_M · 16.5 GB
5
NVIDIA Nemotron 3.5 Lightning 30B A3B BF16NVIDIA
Q4_K_M · 19.3 GB
67.2 tok/sQ4_K_M · 19.3 GB
See all 316 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 (36 GB)

The Mac Studio M5 Max is the fastest 36 GB machine: it runs 27–35B models well. macOS keeps about 3.5 GB, leaving 32.5 GB for models. With 54 TFLOPS it reads long prompts faster than most 36 GB machines: a 4K-token prompt takes 21 s. M5 Max generation, released 2026.

Speed in 36 GB#1460 GB/s
Prompt reading#154 TFLOPS
For models32.5 GBmacOS keeps 3.5 GB
ChipM5 Maxreleased 2026

Mac Studio M5 Max strengths and weaknesses

Strengths

  • Runs 35B MoE models: Qwen3.6 35B A3B at 58.8 tok/s in 22 GB
  • Room for long context: Qwen3.8 27B at Q4 uses 17.9 GB, leaving 14.6 GB free

Weaknesses

  • 70B models don't fit: Llama 3.3 70B Instruct needs 51.3 GB at Q4 vs 32.5 GB usable here
  • Slow prompt reading: A 32K-token prompt on Qwen3.8 27B 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

32.5 GB of 36 GB
9B6.6 GB14B11.9 GB27–31B17.9 GB35B MoE22 GB
14B11.9 GB35B MoE22 GB
For models 32.5 GBmacOS keeps 3.5 GBModel size at Q4

70B models need 51.3 GB at Q4, more than the 32.5 GB available. The 48 GB config fits Mixtral 8x7B Instruct v0.1 →

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, 12 GB more memoryMac Studio M5 Max (48 GB)Apple · 44.5 GB for models · M5 Max
▲ 15% speed▲ 12 GBno price
Same familySame chip, 28 GB more memoryMac Studio M5 Max (64 GB)Apple · 60.5 GB for models · M5 Max
▲ 15% speed▲ 28 GBno price
Same familyMachine, 6% slowerMac Studio M4 Max (36 GB)Apple · 32.5 GB for models · M4 Max
▼ 6% speedsame memoryabout $2,499
Machine, 60% slowerMac Mini M4 (32 GB)Apple · 28.5 GB for models · M4
▼ 60% speed▼ 4 GBabout $999
Laptop, 51% slowerMacBook Pro 14 M5Apple · 24 GB for models · M5
▼ 51% speed▼ 8.5 GBabout $1,839
Reads prompts 9× faster, 32 GB for modelsRTX 5090NVIDIA · 32 GB VRAM · GPU
▲ 9× prompts▲ 263% speedsame memoryabout $5,376rent $0.64/hr

Mac Studio M5 Max tokens per second by model size

Output speed depends on memory bandwidth (460 GB/s). Grey rows need more than the 32.5 GB available.

  1. 9BQwen3.5 9B · Q4_K_M · 6.6 GB
    50.6tok/s
  2. 14BQwen3 14B · Q4_K_M · 11.9 GB
    33.9tok/s
  3. 27–31BQwen3.8 27B · Q4_K_M · 17.9 GB
    18.5tok/s
  4. 35B MoEQwen3.6 35B A3B · Q4_K_M · 22.0 GB
    58.8tok/s
  5. 70BLlama 3.3 70B Instruct · does not fit
    needs 51.3 GB
  6. 120B MoEGPT OSS 120B · does not fit
    needs 70.8 GB

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×M1 Max2022M2 Max2023M3 Max2023M4 Max2025M5 Max2026M1 Max · 15.6 TFLOPS FP16M1 Max · 54.5 tok/s · 400 GB/sM2 Max · 21.5 TFLOPS FP16M2 Max · 54.5 tok/s · 400 GB/sM3 Max · 21.5 TFLOPS FP16M3 Max · 45.9 tok/s · 300 GB/sM4 Max · 25.8 TFLOPS FP16M4 Max · 55.2 tok/s · 409.6 GB/sM5 Max · 54.4 TFLOPS FP16M5 Max · 58.8 tok/s · 460 GB/sCompute 3.5×Writing +8%
1×2×3×4×M12022M22023M32023M42025M52026M1 Max · 15.6 TFLOPS FP16M1 Max · 54.5 tok/s · 400 GB/sM2 Max · 21.5 TFLOPS FP16M2 Max · 54.5 tok/s · 400 GB/sM3 Max · 21.5 TFLOPS FP16M3 Max · 45.9 tok/s · 300 GB/sM4 Max · 25.8 TFLOPS FP16M4 Max · 55.2 tok/s · 409.6 GB/sM5 Max · 54.4 TFLOPS FP16M5 Max · 58.8 tok/s · 460 GB/sCompute 3.5×Writing +8%

Writing speed follows memory bandwidth and barely moved: 54.5 to 58.8 tok/s since the M1 Max. Compute, which sets how fast prompts are read, grew 3.5×. Indexed to the M1 Max; speeds on Qwen3.6 35B A3B, same 32 GB.

All 1115 models on the Mac Studio M5 Max

Featured only

Best quant that fits in 32.5 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 (36 GB) — ranked by Score, recency and Fit
ModelVRAMScoreFit
GLM 4.5 Air
110.5B·IQ2_XXS·45.1 t/s·131K ctx
31.7 GBC44, Usable
Qwen2.5 Coder 3B
3.1B·Q4_K_M·144.4 t/s·33K ctx
2.8 GBS96, Runs great
Llama 2 7B HF
6.7B·Q4_K_M·59.4 t/s·4K ctx
6.5 GBA84, Runs well
Qwen 14B
14.2B·Q4_K_M·34.4 t/s·8K ctx
10.6 GBA73, Runs well
Qwen2.5 14B
14.8B·Q4_K_M·33.7 t/s·131K ctx
12.4 GBA72, Runs well
Yi 34B
34.4B·Q4_K_M·15.0 t/s·4K ctx
21.9 GBB55, Decent
Phi 3 Vision 128k Instruct
4.1B·BF16·34.3 t/s·131K ctx
15.0 GBA73, Runs well
Qwen2 1.5B
1.5B·Q4_K_M·251.6 t/s·131K ctx
1.7 GBS100, Runs great
Llama 3.1 70B LatamGPT SFT 1.0
70.6B·Q2_K·10.4 t/s·4K ctx
31.6 GBD22, Barely runs
Llama 3.1 Nemotron 70B Instruct HF
70.6B·IQ2_M·13.0 t/s·131K ctx
29.5 GBC36, Usable
Baichuan2 13B Base
13B·Q4_K_M·37.5 t/s
11.5 GBA74, Runs well
Codegemma 7B IT
8.5B·Q4_K_M·57.2 t/s
8.2 GBA82, Runs well
Deepseek Llm 7B Base
7B·Q4_K_M·58.4 t/s·4K ctx
6.5 GBA83, Runs well
Mixtral 8x7B v0.1
46.7B·Q4_K_M·32.4 t/s·33K ctx
30.5 GBB51, Decent
Hermes 2 Pro Llama 3 8B
8.0B·Q4_K_M·59.7 t/s·8K ctx
6.2 GBA84, Runs well

Mac Studio M5 Max specs

Specs

Chip
M5 Max
CPU cores
18
GPU cores
32
Unified memory
36 GB
For models
32.5 GB (macOS keeps 3.5 GB)
Memory bandwidth
460 GB/s
FP16 (dense)
54.4 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 (36 GB)?

About 32.5 GB of the 36 GB. macOS keeps roughly 3.5 GB for itself and other apps.

What is the largest model the Mac Studio M5 Max (36 GB) can run?

Qwen3.6 35B A3B (36.0B) at Q4_K_M uses 22.0 GB, 68% of the available memory, and runs at about 58.8 tok/s.

Can the Mac Studio M5 Max (36 GB) run Qwen3.8 27B?

Yes. Qwen3.8 27B fits at Q4_K_M, using 17.9 GB of the 32.5 GB available, and it is the best model that runs well on this machine.

Can the Mac Studio M5 Max (36 GB) run 70B models?

No. Llama 3.3 70B Instruct needs 51.3 GB at Q4, more than the 32.5 GB available.

How long does the Mac Studio M5 Max (36 GB) take to read a long prompt?

On Qwen3.8 27B, 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 Mac Studio M5 Max?

An A40 (48 GB) rents from $0.35/hr.

24 GB GPUsAll MacsDesktops & mini PCsMacBook Pro 14" M4 MaxMacBook Pro 16" M4 MaxMacBook Pro 14" M3 MaxMacBook Pro 16" M4 MaxBest coding LLMsAll hardware