Can Mistral Small 3.2 24B Instruct 2506 run on Mac Studio M2 Ultra 128GB?

YES — Runs Great

C46Usable

Estimated from fit model

Mistral Small 3.2 24B Instruct 2506 needs ~32.2 GB VRAM. Mac Studio M2 Ultra 128GB has 92.2 GB. With Q4_K_M quantization, expect ~32 tok/s.

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

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) — 32.2 GB, 31.7 tok/s, Runs well

32.2 GB required92.2 GB available

35% VRAM used

Fit status

Runs well

Decode

31.7 tok/s

TTFT

6108 ms

Safe context

357K

Memory

32.2 GB / 92.2 GB

Memory breakdown

Weights14.6 GB

KV Cache2.8 GB

Runtime0.9 GB

Headroom13.8 GB

See how fast it feels

See how fast it feelsMistral Small 3.2 24B Instruct 2506 on Mac Studio M2 Ultra 128GB
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: 31.7 tok/s decode · 6.1s TTFT (warm) · 79 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

Workload	Grade	Fit	Decode	TTFT	Context
Chat	C	Runs well	31.7 tok/s	3332 ms	357K
Coding	C	Runs well	31.7 tok/s	6108 ms	357K
Agentic Coding	C	Runs well	31.7 tok/s	8885 ms	357K
Reasoning	C	Runs well	31.7 tok/s	7219 ms	357K
RAG	C	Runs well	31.7 tok/s	11106 ms	357K

Inference speed

Mistral Small 3.2 24B Instruct 2506 inference speed — tokens per second by GPU & Mac

Estimated decode speed (tokens/sec) for Mistral Small 3.2 24B Instruct 2506 at Q4_K_M across popular GPUs and Apple Silicon, using the fastest local runtime per device. Fastest is RTX 5090 32GB at ~82 tok/s. Speed is memory-bandwidth bound, so cards that fit the whole model in VRAM run far faster than ones that offload to system RAM.

GPU / Mac	Memory	Quant	Speed (tok/s)	Fits?
RTX 5090 32GB	32 GB	Q4_K_M	82.0	Fits
RTX 4090 24GB	24 GB	Q4_K_M	52.3	Tight
RX 7900 XTX 24GB	24 GB	Q4_K_M	47.2	Tight
RTX 3090 24GB	24 GB	Q4_K_M	44.8	Tight
Mac Studio M3 Ultra 256GB	256 GB	Q4_K_M	38.0	Fits
MacBook Pro M4 Max 128GB	128 GB	Q4_K_M	34.2	Fits
MacBook Pro M4 Max 64GB	64 GB	Q4_K_M	34.2	Fits
Mac Studio M2 Ultra 128GB	128 GB	Q4_K_M	31.7	Fits
Mac Studio M1 Ultra 128GB	128 GB	Q4_K_M	30.1	Fits
MacBook Pro M4 Pro 48GB	48 GB	Q4_K_M	21.5	Fits
RTX 4080 Super 16GB	16 GB	Q4_K_M	19.1	Too big
MacBook Pro M3 Max 64GB	64 GB	Q4_K_M	16.4	Fits
MacBook Pro M1 Max 64GB	64 GB	Q4_K_M	15.0	Fits
RTX 4070 12GB	12 GB	Q4_K_M	6.7	Too big
RTX 3060 12GB	12 GB	Q4_K_M	4.2	Too big
RTX 4060 8GB	8 GB	Q4_K_M	2.0	Too big

Estimates for single-stream decoding at Q4_K_M; real tokens/sec varies with prompt length, context, batch size, and runtime build. Prompt processing (prefill) is faster than the decode figures shown here.

Quantization options

How Mistral Small 3.2 24B Instruct 2506 (24B params) fits at each quantization level on Mac Studio M2 Ultra 128GB (92.2 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	9.4 GB	Low	D40
Q3_K_S	3	11.8 GB	Low	D40
NVFP4	4	13.4 GB	Medium	C40
Q4_K_M	4	14.6 GB	Medium	C40
Q5_K_M	5	17.3 GB	High	C41
Q6_K	6	19.7 GB	High	C41
Q8_0	8	25.7 GB	Very High	C42
F16Best for your GPU	16	49.2 GB	Maximum	C47

Get started

Copy-paste commands to run Mistral Small 3.2 24B Instruct 2506 on your machine.

Run

lms load hf-unsloth--mistral-small-3-2-24b-instruct-2506-gguf && lms server start

Opções de upgrade

Hardware que roda bem Mistral Small 3.2 24B Instruct 2506

RTX PRO 6000 Blackwell Workstation Edition 96GBOpção econômica

1792 GB/s (+992)

Raises estimated decode speed by about 224%.102.8 tok/s decodificação

Raises estimated decode speed by about 224%.

~$9,999 MSRP

RTX PRO 6000 Blackwell Server Edition 96GBMelhor custo-benefício

1597 GB/s (+797)

Raises estimated decode speed by about 189%.91.6 tok/s decodificação

Raises estimated decode speed by about 189%.

~$9,999 MSRP

Frequently asked questions

Can Mac Studio M2 Ultra 128GB run Mistral Small 3.2 24B Instruct 2506?

Yes, Mac Studio M2 Ultra 128GB can run Mistral Small 3.2 24B Instruct 2506 with a C grade (Runs well). Expected decode speed: 31.7 tok/s.

How much VRAM does Mistral Small 3.2 24B Instruct 2506 need?

Mistral Small 3.2 24B Instruct 2506 (24B parameters) requires approximately 32.2 GB of memory with Q4_K_M quantization.

What is the best quantization for Mistral Small 3.2 24B Instruct 2506?

The recommended quantization for Mistral Small 3.2 24B Instruct 2506 is Q4_K_M, which balances quality and memory efficiency.

What speed will Mistral Small 3.2 24B Instruct 2506 run at on Mac Studio M2 Ultra 128GB?

On Mac Studio M2 Ultra 128GB, Mistral Small 3.2 24B Instruct 2506 achieves approximately 31.7 tokens per second decode speed with a time-to-first-token of 6108ms using Q4_K_M quantization.

Can Mac Studio M2 Ultra 128GB run Mistral Small 3.2 24B Instruct 2506 for coding?

For coding workloads, Mistral Small 3.2 24B Instruct 2506 on Mac Studio M2 Ultra 128GB receives a C grade with 31.7 tok/s and 357K context.

What context window can Mistral Small 3.2 24B Instruct 2506 use on Mac Studio M2 Ultra 128GB?

On Mac Studio M2 Ultra 128GB, Mistral Small 3.2 24B Instruct 2506 can safely use up to 357K tokens of context. The model's official context limit is —, but available memory constrains the safe maximum.

Is unified memory on Mac Studio M2 Ultra 128GB as fast as VRAM for Mistral Small 3.2 24B Instruct 2506?

Not always. Mac Studio M2 Ultra 128GB can often fit larger models thanks to unified memory, but a discrete GPU with dedicated high-bandwidth VRAM may still decode faster once the model fits. For this combination, the important distinction is capacity versus sustained throughput.

See all results for Mac Studio M2 Ultra 128GB See all hardware for Mistral Small 3.2 24B Instruct 2506

Embed this result▼

Paste this snippet into any page to show a live fit card.

<iframe src="https://willitrunai.com/embed/hf-unsloth--mistral-small-3-2-24b-instruct-2506-gguf-on-m2-ultra-128gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview: