GLM-4 32B 0414 on Apple M2 Ultra (192 GB)

Can the Apple M2 Ultra (192 GB) run GLM-4 32B 0414?

Yes, it runs well

Runs well. At Q4_K_M it needs about 32.5 GB of 144 GB usable, with an estimated speed of 26.0 tok/s.

Best quant: Q4_K_M
Generation speed: 26.0 tok/s
Prompt processing: 518 tok/s
First token (1K prompt): 2.0 s
Largest safe context: 32K

Best app to run it on the Apple M2 Ultra (192 GB)

On macOS

LM Studio

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

Runs the MLX build, typically up to 10% faster than GGUF for a model with 32.6B active parameters on Apple silicon.
Point-and-click: search, download and chat in one window.

Open GLM-4 32B 0414 in LM Studio

llama.cpp Usually the first to support new model architectures.
Ollama Runs the Hugging Face GGUF with ollama run hf.co/…

Want an agent? GLM-4 32B 0414 supports tool use, so you can pair a llama.cpp or MLX LM server with one of these. Apps that run on this computer come first.

Coding agents

Personal assistants

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 GLM-4 32B 0414

Recommended for the Apple M2 Ultra (192 GB): Q4_K_M. 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 Q4_K_M file (opens in a new tab)

huggingface.co/unsloth/GLM-4-32B-0414-GGUF (opens in a new tab)

ollama run hf.co/unsloth/GLM-4-32B-0414-GGUF:Q4_K_M

Original weights · Hugging Face

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

huggingface.co/zai-org/GLM-4-32B-0414 (opens in a new tab)

No sign-in needed.

Official repo · GitHub

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

github.com/zai-org/GLM-4 (opens in a new tab)

git clone https://github.com/zai-org/GLM-4

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

How it’s rated

Quality 70Great quality (A) · 70 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking.: How capable the model itself is at Q4_K_M, 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 well: How well it runs on the Apple M2 Ultra (192 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.	31.6 / 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.	12.1 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.	5.2 / 15
Match score	73.9 / 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 M2 Ultra (192 GB)

Quant	Memory needed	Available	Speed	Runs	Quality	Download
`Q2_K`	25.7 GB	144 GB	42.6 tok/s	Runs great	51Fair quality (C) · 51 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q3_K_M`	28.6 GB	144 GB	33.6 tok/s	Runs great	62Good quality (B) · 62 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q4_K_M`	32.5 GB	144 GB	26.0 tok/s	Runs well	70Great quality (A) · 70 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q5_K_M`	35.8 GB	144 GB	21.4 tok/s	Runs well	71Great quality (A) · 71 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q6_K`	39.3 GB	144 GB	18.0 tok/s	Runs well	73Great quality (A) · 73 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q8_0`	47.2 GB	144 GB	13.2 tok/s	Runs well	74Great quality (A) · 74 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`F16`	77.8 GB	144 GB	6.1 tok/s	Slow	74Great quality (A) · 74 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.	GGUF · folder (opens in a new tab) 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

71Great quality (A) · 71 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 5090 Q5_K_M · 55.8 tok/s · Runs great
70Great quality (A) · 70 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 4090 Q4_K_M · 38.0 tok/s · Runs well (tight fit)
70Great quality (A) · 70 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 3090 Q4_K_M · 35.3 tok/s · Runs well (tight fit)
70Great quality (A) · 70 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. AMD Radeon RX 7900 XTX Q4_K_M · 31.2 tok/s · Runs well (tight fit)
73Great quality (A) · 73 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA RTX PRO 6000 Blackwell Q6_K · 46.9 tok/s · Runs great
70Great quality (A) · 70 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 Max (64 GB) Q4_K_M · 17.7 tok/s · Runs well

Quick answers

How much memory does GLM-4 32B 0414 need on the M2 Ultra 192GB?

At Q4_K_M it needs about 32.5 GB. The M2 Ultra 192GB has 144 GB usable.

How fast is GLM-4 32B 0414 on the M2 Ultra 192GB?

About 26.0 tok/s at Q4_K_M, an estimate from the hardware’s memory bandwidth and the size of the model.

Which quantization of GLM-4 32B 0414 should I use on the M2 Ultra 192GB?

Q4_K_M. 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.