willitrun·ai

Can WizardLM 13B run on Mac Studio M2 Ultra 128GB?

YES — Runs Great

B69Good
Estimated from fit model

WizardLM 13B needs ~34.9 GB VRAM. Mac Studio M2 Ultra 128GB has 92.2 GB. With Q4_K_M quantization, expect ~59 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) — 34.9 GB, 58.5 tok/s, Runs well
34.9 GB required92.2 GB available
38% VRAM used

Fit status

Runs well

Decode

58.5 tok/s

TTFT

3309 ms

Safe context

8K

Memory

34.9 GB / 92.2 GB

Memory breakdown

Weights7.9 GB
KV Cache12.2 GB
Runtime0.9 GB
Headroom13.8 GB

See how fast it feels

See how fast it feelsWizardLM 13B on Mac Studio M2 Ultra 128GB
1st promptCold start — includes initialization
>Solve: If a train travels at 60 mph for 2.5 hours, how far does it go?
Let me work through this step by step. Given information: - Speed = 60 mph (miles per hour) - Time = 2.5 hours Formula: Distance = Speed x Time Calculation: Distance = 60 mph x 2.5 h = 150 miles The train travels 150 miles. Double-check: at 60 mph the train covers 60 miles each hour. In half an hour it covers 30 miles. So in 2.5 hours: 60 + 60 + 30 = 150 miles. Confirmed.
2nd promptWarm — model ready, no init overhead
>What if the train stops for 20 minutes halfway?
The stop doesn't change the total distance covered while moving, but it affects total trip time. Breakdown: - First half: 1.25 hours at 60 mph = 75 miles - Stop: 20 minutes (no distance) - Second half: 1.25 hours at 60 mph = 75 miles Total distance: 150 miles (same) Total time: 2.5 h + 0.33 h = 2.83 hours Average speed: 150 / 2.83 = 53 mph The distance stays the same but average speed drops to 53 mph because of the stop.
Estimated: 58.5 tok/s decode · 3.3s TTFT (warm) · 146 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
ChatBRuns well58.5 tok/s1805 ms8K
CodingBRuns well58.5 tok/s3309 ms8K
Agentic CodingARuns well58.5 tok/s4813 ms8K
ReasoningBRuns well58.5 tok/s3910 ms8K
RAGARuns well58.5 tok/s6016 ms8K

Inference speed

WizardLM 13B inference speed — tokens per second by GPU & Mac

Estimated decode speed (tokens/sec) for WizardLM 13B at Q4_K_M across popular GPUs and Apple Silicon, using the fastest local runtime per device. Fastest is RTX 5090 32GB at ~151 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 / MacMemoryQuantSpeed (tok/s)Fits?
NVIDIARTX 5090 32GB
32 GBQ4_K_M151.4Fits
NVIDIARTX 4090 24GB
24 GBQ4_K_M96.6Offloads
RX 7900 XTX 24GB
24 GBQ4_K_M87.2Offloads
NVIDIARTX 3090 24GB
24 GBQ4_K_M82.6Offloads
Mac Studio M3 Ultra 256GB
256 GBQ4_K_M70.2Fits
Mac Studio M2 Ultra 128GB
128 GBQ4_K_M58.5Fits
Mac Studio M1 Ultra 128GB
128 GBQ4_K_M55.5Fits
MacBook Pro M4 Max 128GB
128 GBQ4_K_M38.2Fits
MacBook Pro M4 Max 64GB
64 GBQ4_K_M38.2Fits
MacBook Pro M3 Max 64GB
64 GBQ4_K_M30.3Fits
MacBook Pro M1 Max 64GB
64 GBQ4_K_M27.7Fits
NVIDIARTX 4080 Super 16GB
16 GBQ4_K_M27.1Too big
MacBook Pro M4 Pro 48GB
48 GBQ4_K_M23.3Fits
NVIDIARTX 4070 12GB
12 GBQ4_K_M9.5Too big
NVIDIARTX 3060 12GB
12 GBQ4_K_M6.0Too big
NVIDIARTX 4060 8GB
8 GBQ4_K_M3.8Too 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 WizardLM 13B (13B params) fits at each quantization level on Mac Studio M2 Ultra 128GB (92.2 GB usable).

QuantBitsVRAMQualityFit
Q2_K
2
5.1 GB
LowB60
Q3_K_S
3
6.4 GB
LowB60
NVFP4
4
7.3 GB
MediumB60
Q4_K_M
4
7.9 GB
MediumB60
Q5_K_M
5
9.4 GB
HighB60
Q6_K
6
10.7 GB
HighB60
Q8_0
8
13.9 GB
Very HighB61
F16Best for your GPU
16
26.7 GB
MaximumB62

Get started

Copy-paste commands to run WizardLM 13B on your machine.

Run

lms load WizardLM-13B-V1.0 && lms server start

Opciones de mejora

Hardware que ejecuta bien WizardLM 13B

NVIDIARTX PRO 6000 Blackwell Workstation Edition 96GBOpción económica
1792 GB/s (+992)
B
Sube la velocidad estimada de decodificación alrededor de un 211%.182 tok/s decodificación

Sube la velocidad estimada de decodificación alrededor de un 211%.

~$9,999 MSRP

NVIDIARTX PRO 6000 Blackwell Server Edition 96GBMejor relación calidad-precio
1597 GB/s (+797)
B
Sube la velocidad estimada de decodificación alrededor de un 189%.169.2 tok/s decodificación

Sube la velocidad estimada de decodificación alrededor de un 189%.

~$9,999 MSRP

Frequently asked questions

Can Mac Studio M2 Ultra 128GB run WizardLM 13B?

Yes, Mac Studio M2 Ultra 128GB can run WizardLM 13B with a B grade (Runs well). Expected decode speed: 58.5 tok/s.

How much VRAM does WizardLM 13B need?

WizardLM 13B (13B parameters) requires approximately 34.9 GB of memory with Q4_K_M quantization.

What is the best quantization for WizardLM 13B?

The recommended quantization for WizardLM 13B is Q4_K_M, which balances quality and memory efficiency.

What speed will WizardLM 13B run at on Mac Studio M2 Ultra 128GB?

On Mac Studio M2 Ultra 128GB, WizardLM 13B achieves approximately 58.5 tokens per second decode speed with a time-to-first-token of 3309ms using Q4_K_M quantization.

Can Mac Studio M2 Ultra 128GB run WizardLM 13B for coding?

For coding workloads, WizardLM 13B on Mac Studio M2 Ultra 128GB receives a B grade with 58.5 tok/s and 8K context.

What context window can WizardLM 13B use on Mac Studio M2 Ultra 128GB?

On Mac Studio M2 Ultra 128GB, WizardLM 13B can safely use up to 8K tokens of context. The model's official context limit is 8K, but available memory constrains the safe maximum.

Is unified memory on Mac Studio M2 Ultra 128GB as fast as VRAM for WizardLM 13B?

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 128GBSee all hardware for WizardLM 13B
Embed this result▼

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

<iframe src="https://willitrunai.com/embed/wizardlm-13b-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: