Phi-3.5 MoE (42B-A6.6B) on Apple M5 Ultra (256 GB)

Can the Apple M5 Ultra (256 GB) run Phi-3.5 MoE (42B-A6.6B)?

Yes, it runs great

Runs great. At Q8_0 it needs about 61.4 GB of 192 GB usable, with an estimated speed of 71.3 tok/s.

Best quant: Q8_0
Generation speed: 71.3 tok/s
Prompt processing: 3,542 tok/s
First token (1K prompt): 0.3 s
Largest safe context: 128K

Best app to run it on the Apple M5 Ultra (256 GB)

On macOS

LM Studio

Desktop app · runs on llama.cpp, MLX · use the MLX build

Runs the MLX build, typically 20–35% faster than GGUF for a model with 6.6B active parameters on Apple silicon.
Point-and-click: search, download and chat in one window.

Open Phi-3.5 MoE (42B-A6.6B) in LM Studio

MLX LM Runs the MLX build, typically 20–35% faster than GGUF for a model with 6.6B active parameters on Apple silicon.
Jan Runs the MLX build, typically 20–35% faster than GGUF for a model with 6.6B active parameters on Apple silicon.

How we choose: the files published for this model, your operating system and GPU, whether it fits fully in memory, and published speed tests (MLX vs llama.cpp on Apple silicon, vLLM vs llama.cpp vs Ollama on NVIDIA). Details.

Download Phi-3.5 MoE (42B-A6.6B)

Recommended for the Apple M5 Ultra (256 GB): Q8_0. Every other quantization has its own download link in the table below.

Quantized files · GGUF

Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.

Download the Q8_0 file (opens in a new tab)

huggingface.co/lmstudio-community/Phi-3.5-MoE-instruct-GGUF (opens in a new tab)

ollama run hf.co/lmstudio-community/Phi-3.5-MoE-instruct-GGUF:Q8_0

Original weights · Hugging Face

Full-precision files from Microsoft, for fine-tuning, vLLM, Transformers.

huggingface.co/microsoft/Phi-3.5-MoE-instruct (opens in a new tab)

No sign-in needed.

Official repo · GitHub

Reference code, inference scripts and docs from Microsoft. The weights themselves download from Hugging Face.

github.com/microsoft/PhiCookBook (opens in a new tab)

Microsoft’s official Phi cookbook: setup, inference and fine-tuning guides.

git clone https://github.com/microsoft/PhiCookBook

Links checked 3 days ago against Hugging Face, Ollama and GitHub.

How it’s rated

Quality 56Good quality (B) · 56 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking.: How capable the model itself is at Q8_0, from its editorial quality tier scaled by the quantization or precision. It’s the same on any hardware that runs this variant. Scale: A+ 85+ · A 70+ · B 55+ · C 40+ · D under 40.
Runs Runs great: How well it runs on the Apple M5 Ultra (256 GB): speed, capped by how well it fits in memory. Runs great: 30+ tokens/s (or near the image, video or decision speed target) · Runs well: 10+ tokens/s · Slow: 3+ tokens/s · Very slow: under 3 · Won’t run: doesn’t fit even with offload. A tight fit or RAM offload caps it at Runs well; heavy offload caps it at Slow.

Match score

To order lists, each result also gets a match score out of 100 from four parts. It favors the most capable model that still runs well.

Model qualityEditorial quality tier of the model, scaled by the quantization or precision used.	25.2 / 45
Memory fitFull marks when the model fits in the share of unified memory the GPU can use; fewer when it goes past that limit.	25.0 / 25
SpeedGeneration speed against a 40 tokens/s target for LLMs, or render time against a target for image and video.	15.0 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.	7.4 / 15
Match score	72.6 / 100

The match score orders results; it is not shown as a grade. Quality and Runs are the two ratings in the tables.

Every quantization on the Apple M5 Ultra (256 GB)

Quant	Memory needed	Available	Speed	Runs	Quality	Download
`Q2_K`	33.8 GB	192 GB	159 tok/s	Runs great	39Basic quality (D) · 39 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking.	GGUF files (opens in a new tab)
`Q3_K_M`	37.6 GB	192 GB	138 tok/s	Runs great	47Fair quality (C) · 47 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (Q3_K_L) (opens in a new tab)
`Q4_K_M`	42.6 GB	192 GB	117 tok/s	Runs great	53Fair quality (C) · 53 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q5_K_M`	46.8 GB	192 GB	102 tok/s	Runs great	54Fair quality (C) · 54 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking.	GGUF files (opens in a new tab)
`Q6_K`	51.4 GB	192 GB	90.0 tok/s	Runs great	55Good quality (B) · 55 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q8_0`	61.4 GB	192 GB	71.3 tok/s	Runs great	56Good quality (B) · 56 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`F16`	101 GB	192 GB	37.7 tok/s	Runs great	56Good quality (B) · 56 of 100: how capable the model is at F16, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking.	Original weights (opens in a new tab)

Memory figures assume 8K of context. Longer conversations or documents grow the KV cache.

Other hardware that runs it

55Good quality (B) · 55 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. Apple M4 Max (64 GB) Q6_K · 49.8 tok/s · Runs great
55Good quality (B) · 55 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. Apple M1 Max (64 GB) Q6_K · 38.3 tok/s · Runs great
53Fair quality (C) · 53 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. Apple M4 Pro (48 GB) Q4_K_M · 38.1 tok/s · Runs great
53Fair quality (C) · 53 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 5090 Q4_K_M · 157 tok/s · Runs well (tight fit)
56Good quality (B) · 56 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA RTX PRO 6000 Blackwell Q8_0 · 103 tok/s · Runs great
55Good quality (B) · 55 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. Apple M4 Max (128 GB) Q6_K · 49.8 tok/s · Runs great

Quick answers

How much memory does Phi-3.5 MoE (42B-A6.6B) need on the M5 Ultra 256GB?

At Q8_0 it needs about 61.4 GB. The M5 Ultra 256GB has 192 GB usable.

How fast is Phi-3.5 MoE (42B-A6.6B) on the M5 Ultra 256GB?

About 71.3 tok/s at Q8_0, an estimate from the hardware’s memory bandwidth and the size of the model.

Which quantization of Phi-3.5 MoE (42B-A6.6B) should I use on the M5 Ultra 256GB?

Q8_0. Of the versions that fit, it is the one that runs best; when two run equally well, the check prefers the higher-rated one.

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.