VOOZH about

URL: https://willitrunai.com/can-run/olmo-2-32b-on-m2-ultra-64gb

⇱ OLMo 2 32B on Mac Studio M2 Ultra 64GB? YES

Can OLMo 2 32B run on Mac Studio M2 Ultra 64GB?

YES — Runs Great

A85Great
○Estimated from fit model

OLMo 2 32B needs ~31.2 GB VRAM. Mac Studio M2 Ultra 64GB has 46.1 GB. With Q4_K_M quantization, expect ~26 tok/s.

Runtime: llama.cppCapacity: RoomyBandwidth: HighStack: StandardBottleneck: Balanced
Share:

Operating mode

Choose the run profile you care about

Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.

Current mode

Balanced

Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 31.2 GB, 25.7 tok/s, Runs well
31.2 GB required46.1 GB available
68% VRAM used

Fit status

Runs well

Decode

25.7 tok/s

TTFT

7541 ms

Safe context

4K

Memory

31.2 GB / 46.1 GB

Memory breakdown

Weights19.5 GB
KV Cache3.9 GB
Runtime0.9 GB
Headroom6.9 GB

See how fast it feels

See how fast it feelsOLMo 2 32B on Mac Studio M2 Ultra 64GB
1st promptCold start — includes initialization
>What is local AI inference?
Local AI inference means running an AI language model directly on your own hardware — your laptop, desktop, or server — instead of sending requests to a remote cloud API. When you run inference locally the model weights are loaded into your GPU or unified memory. Each token you generate requires reading those weights from memory, so memory bandwidth is the main bottleneck for decode speed. Key benefits of running locally: - Full privacy: your prompts never leave your machine - No per-token cost or rate limits - Works offline once the model is downloaded - Latency depends only on your hardware
2nd promptWarm — model ready, no init overhead
>How much VRAM do I need?
It depends on the model size and quantization level. A rough rule of thumb: Model size Q4 (4-bit) Q8 (8-bit) FP16 7B params ~4.3 GB ~7.5 GB ~14 GB 13B params ~7.9 GB ~13.9 GB ~26 GB 70B params ~42.7 GB ~74.9 GB ~140 GB Most people use 4-bit quantization (Q4_K_M) which gives 90-95% of full quality at a fraction of the memory. A 24 GB GPU can comfortably run most 7B-13B models.
Estimated: 25.7 tok/s decode · 7.5s TTFT (warm) · 64 tok/s prefill

What limits this setup

This setup is broadly balanced for this model.

Shared-memory contention still exists

The OS, browser, and inference runtime all compete for the same physical memory pool, so real-world headroom is less forgiving than raw capacity suggests.

Best improvement path

Performance by workload

WorkloadGradeFitDecodeTTFTContext
ChatARuns well25.7 tok/s4113 ms4K
CodingARuns well25.7 tok/s7541 ms4K
Agentic CodingSRuns well25.7 tok/s10969 ms4K
ReasoningARuns well25.7 tok/s8912 ms4K
RAGSRuns well25.7 tok/s13711 ms4K

Quantization options

How OLMo 2 32B (32B params) fits at each quantization level on Mac Studio M2 Ultra 64GB (46.1 GB usable).

QuantBitsVRAMQualityFit
Q2_K
2
12.5 GB
LowA76
Q3_K_S
3
15.7 GB
LowA78
NVFP4
4
17.9 GB
MediumA78
Q4_K_M
4
19.5 GB
MediumA79
Q5_K_M
5
23.0 GB
HighA80
Q6_K
6
26.2 GB
HighA81
Q8_0Best for your GPU
8
34.2 GB
Very HighA80
F16
16
65.6 GB
MaximumF0

Get started

Copy-paste commands to run OLMo 2 32B on your machine.

Run

lms load OLMo-2-0325-32B-Instruct && lms server start

Your hardware

More models your Mac Studio M2 Ultra 64GB can run

ModelParamsGradeDecodeCapabilities
👁 Alibaba
Qwen 3.6 35B A3B		35BS59 tok/s
👁 Alibaba
Qwen 3.5 35B A3B		35BS64.1 tok/s
👁 Moonshot AI
Kimi Linear 48B A3B		48BA15.8 tok/s

Frequently asked questions

See all results for Mac Studio M2 Ultra 64GBSee all hardware for OLMo 2 32B