Is the M5 Ultra faster than the M4 Max for LLMs?

The M5 Ultra reads memory about 120% faster than the M4 Max (1,200 vs 546 GB/s), so it writes about 120% faster with any model that fits both. Reading a long prompt depends on GPU compute instead, which this comparison doesn’t cover.

Which can run bigger models, the M4 Max or the M5 Ultra?

The M5 Ultra: The M5 Ultra goes up to 512 GB of memory, the M4 Max to 128 GB. At its largest it runs 61 of 73 open models well, against 47.

M4 Max vs M5 Ultra for local LLMs: memory, speed, which to buy

Side by side

	M4 Max	M5 Ultra
Memory sizes	36, 48, 64, 128 GB	96, 256, 512 GB
Memory bandwidth	410–546 GB/s	1,200 GB/s
For models, at most	about 96 GB	about 384 GB
Models that run well	47 of 73	61 of 73
MLX seat class	Pro	Ultra Latest
Found in	MacBook Pro (14-inch, M4 Pro or M4 Max, 2024), MacBook Pro (16-inch, 2024), Mac Studio (2025)	Mac Studio (M5 Max or M5 Ultra)

Speed with popular models

Tokens a second for one conversation, each chip at its largest memory size (128 GB and 512 GB), with the best version of the model that fits.

Model	M4 Max	M5 Ultra
Llama 3.1 8B Instruct8B	29–51 8-bit	64–113 8-bit
gpt-oss-20b21B · 3.6B active	129–230 native 4.25-bit	284–505 native 4.25-bit
Qwen3.6 35B A3B35B · 3B active	77–137 8-bit	169–300 8-bit
Qwen3.8 27B27B	11–20 6-bit	19–33 8-bit
Llama 3.3 70B Instruct70B	3–6 8-bit	14–24 4-bit
gpt-oss-120b117B · 5.1B active	91–162 native 4.25-bit	200–356 native 4.25-bit
Qwen3.5 122B A10B122B · 10B active	44–78 4-bit	51–91 8-bit
DeepSeek V4 Flash 0731284B · 13B active	60–107 2-bit lower quality	39–70 8-bit

Questions

Is the M5 Ultra faster than the M4 Max for LLMs?: The M5 Ultra reads memory about 120% faster than the M4 Max (1,200 vs 546 GB/s), so it writes about 120% faster with any model that fits both. Reading a long prompt depends on GPU compute instead, which this comparison doesn’t cover.
Which can run bigger models, the M4 Max or the M5 Ultra?: The M5 Ultra: The M5 Ultra goes up to 512 GB of memory, the M4 Max to 128 GB. At its largest it runs 61 of 73 open models well, against 47.

More comparisons

The M4 Max for local AI · The M5 Ultra for local AI · Compare every model · The best Mac for local LLMs

How it’s worked out

Memory: the model’s parameters at 8, 6, 4, 3 or 2 bits per weight plus MLX’s group scales (half a bit per weight), or its native format, with room for the context on top. By default macOS lets the GPU use about 75% of unified memory. Speed: memory bandwidth divided by the bytes read per token is the ceiling for one conversation; in llama.cpp’s public Apple Silicon benchmarks real runs reach about 45% to 80% of it, which is the range shown. These are estimates, not measurements: measure your own Mac on MLX Bench. Model figures from each official model card, chip figures from Apple’s Tech Specs. Checked 2026-10-19. Same data as mlx.app/compare.