Can Granite Code 20B run on MacBook Pro M3 Pro 18GB?

YES — With Q2_K

B67Good

Estimated from fit model

Granite Code 20B needs ~13.8 GB VRAM. MacBook Pro M3 Pro 18GB has 13.0 GB. With Q2_K quantization, expect ~12 tok/s.

Runtime: llama.cppCapacity: OffloadBandwidth: Very lowStack: StandardBottleneck: Host offload

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.

Granite Code 20B at Q4_K_M needs 18.2 GB — too much for MacBook Pro M3 Pro 18GB (13.0 GB). Runs at Q2_K (13.8 GB) with low quality.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 18.2 GB, exceeds 13.0 GB available

18.2 GB required13.0 GB available

140% VRAM needed

5.2 GB over capacity — needs offload or smaller quantization

Fit status

Too heavy

Decode

6.1 tok/s

TTFT

31704 ms

Safe context

Memory

18.2 GB / 13.0 GB

Offload

30%

Memory breakdown

Weights12.2 GB

KV Cache3.2 GB

Runtime0.9 GB

Headroom1.9 GB

See how fast it feels

With memory offload — actual speed may be lower

See how fast it feelsGranite Code 20B on MacBook Pro M3 Pro 18GB
1st promptCold start — includes initialization
>Write a Python function to sort a list
Here's a Python function that sorts a list using quicksort:

def quicksort(arr):
    if len(arr) <= 1:
        return arr
    pivot = arr[len(arr) // 2]
    left   = [x for x in arr if x < pivot]
    middle = [x for x in arr if x == pivot]
    right  = [x for x in arr if x > pivot]
    return quicksort(left) + middle + quicksort(right)

# Example usage
numbers = [3, 6, 8, 10, 1, 2, 1]
print(quicksort(numbers))  # [1, 1, 2, 3, 6, 8, 10]

The algorithm picks a pivot, partitions the list into smaller, equal, and larger elements, then recursively sorts each partition and concatenates the results.
2nd promptWarm — model ready, no init overhead
>Now add type hints and a docstring
def quicksort(arr: list[int]) -> list[int]:
    """Sort a list using the quicksort algorithm.

    Args:
        arr: List of integers to sort.

    Returns:
        A new sorted list.
    """
    if len(arr) <= 1:
        return arr
    pivot = arr[len(arr) // 2]
    left   = [x for x in arr if x < pivot]
    middle = [x for x in arr if x == pivot]
    right  = [x for x in arr if x > pivot]
    return quicksort(left) + middle + quicksort(right)

Added type hints for the input and return type, plus a Google-style docstring.
Estimated: 6.1 tok/s decode · 31.7s TTFT (warm) · 15 tok/s prefill

What limits this setup

It fits through host-memory offload, and offload is the main reason performance drops.

CPU or host-memory offload is active

About 10% of the working set spills out of accelerator memory, which usually hurts latency and sustained decode throughput.

Very little memory headroom

You can run the model, but there is not much room left for longer context, bigger batches, extra apps, or future model updates.

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

Remove offload with more accelerator memory

Prioritize a GPU or unified-memory tier that fits the whole model natively. Removing offload usually helps more than small compute gains.

Buy headroom, not only minimum fit

A slightly larger memory tier gives you safer context growth and makes the recommendation more future-proof.

Increase host RAM if you keep offloading

This setup may need roughly 0.5 GB of extra host RAM just for the offloaded portion, before OS and other tools.

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Chat	F	Too heavy	6.8 tok/s	15532 ms	4K
Coding	F	Too heavy	6.1 tok/s	31704 ms	4K
Agentic Coding	F	Too heavy	5.1 tok/s	55316 ms	4K
Reasoning	F	Too heavy	6.1 tok/s	37468 ms	4K
RAG	F	Too heavy	5.1 tok/s	69145 ms	4K

Quantization options

How Granite Code 20B (20B params) fits at each quantization level on MacBook Pro M3 Pro 18GB (13.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_KBest for your GPU	2	7.8 GB	Low	A81
Q3_K_S	3	9.8 GB	Low	F0
NVFP4	4	11.2 GB	Medium	F0
Q4_K_M	4	12.2 GB	Medium	F0
Q5_K_M	5	14.4 GB	High	F0
Q6_K	6	16.4 GB	High	F0
Q8_0	8	21.4 GB	Very High	F0
F16	16	41.0 GB	Maximum	F0

Get started

Copy-paste commands to run Granite Code 20B on your machine.

Run

ollama run granite-code:20b

Upgrade options

Hardware that runs Granite Code 20B well

MacBook Pro M4 32GBBudget pick

32 GB Unified (+14)

Makes the model fit on the accelerator instead of staying completely out of reach.9.9 tok/s decode

Makes the model fit on the accelerator instead of staying completely out of reach.

Removes host-memory offload, which is usually the single biggest latency and throughput win.

~$799 MSRP

Mac mini M4 32GBBest value

32 GB Unified (+14)

Makes the model fit on the accelerator instead of staying completely out of reach.9.9 tok/s decode

Makes the model fit on the accelerator instead of staying completely out of reach.

Removes host-memory offload, which is usually the single biggest latency and throughput win.

~$1,099 MSRP

Mac mini M4 64GBApple upgrade

64 GB Unified (+46)

Makes the model fit on the accelerator instead of staying completely out of reach.9.9 tok/s decode

Makes the model fit on the accelerator instead of staying completely out of reach.

Removes host-memory offload, which is usually the single biggest latency and throughput win.

~$1,099 MSRP

RTX 5090 Laptop 24GBBiggest leap

24 GB VRAM (+6)896 GB/s (+746)

Makes the model fit on the accelerator instead of staying completely out of reach.66.6 tok/s decode

Makes the model fit on the accelerator instead of staying completely out of reach.

Removes host-memory offload, which is usually the single biggest latency and throughput win.

Frequently asked questions

Can MacBook Pro M3 Pro 18GB run Granite Code 20B?

Yes, MacBook Pro M3 Pro 18GB can run Granite Code 20B at Q2_K quantization (Runs with offload (needs ~0.5 GB host RAM)). The recommended Q4_K_M requires 18.2 GB which exceeds available memory, but at Q2_K it needs only 13.8 GB. Expected decode speed: 11.5 tok/s.

How much VRAM does Granite Code 20B need?

Granite Code 20B (20B parameters) requires approximately 18.2 GB at Q4_K_M quantization. On MacBook Pro M3 Pro 18GB, it fits at Q2_K using 13.8 GB.

What is the best quantization for Granite Code 20B?

The recommended quantization is Q4_K_M, but on MacBook Pro M3 Pro 18GB the best fitting quantization is Q2_K, which uses 13.8 GB.

What speed will Granite Code 20B run at on MacBook Pro M3 Pro 18GB?

On MacBook Pro M3 Pro 18GB, Granite Code 20B achieves approximately 11.5 tokens per second decode speed with a time-to-first-token of 16846ms using Q2_K quantization.

Can MacBook Pro M3 Pro 18GB run Granite Code 20B for coding?

For coding workloads, Granite Code 20B on MacBook Pro M3 Pro 18GB receives a F grade with 6.1 tok/s and 4K context.

What context window can Granite Code 20B use on MacBook Pro M3 Pro 18GB?

On MacBook Pro M3 Pro 18GB, Granite Code 20B can safely use up to 8K tokens of context at Q2_K quantization. The model's official context limit is 8K, but available memory constrains the safe maximum.

What should I upgrade first if Granite Code 20B feels slow on MacBook Pro M3 Pro 18GB?

Remove offload with more accelerator memory. Prioritize a GPU or unified-memory tier that fits the whole model natively. Removing offload usually helps more than small compute gains.

Is unified memory on MacBook Pro M3 Pro 18GB as fast as VRAM for Granite Code 20B?

Not always. MacBook Pro M3 Pro 18GB 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 MacBook Pro M3 Pro 18GB See all hardware for Granite Code 20B

Embed this result▼

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

<iframe src="https://willitrunai.com/embed/granite-code-20b-on-m3-pro-18gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview:

What limits this setup

It fits through host-memory offload, and offload is the main reason performance drops.

CPU or host-memory offload is active

About 10% of the working set spills out of accelerator memory, which usually hurts latency and sustained decode throughput.

Very little memory headroom

You can run the model, but there is not much room left for longer context, bigger batches, extra apps, or future model updates.

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

Remove offload with more accelerator memory

Prioritize a GPU or unified-memory tier that fits the whole model natively. Removing offload usually helps more than small compute gains.

Buy headroom, not only minimum fit

A slightly larger memory tier gives you safer context growth and makes the recommendation more future-proof.

Increase host RAM if you keep offloading

This setup may need roughly 0.5 GB of extra host RAM just for the offloaded portion, before OS and other tools.

Workload

Grade

Fit

Decode

TTFT

Context

Chat

Too heavy

6.8 tok/s

15532 ms

Coding

Too heavy

6.1 tok/s

31704 ms

Agentic Coding

Too heavy

5.1 tok/s

55316 ms

Reasoning

Too heavy

6.1 tok/s

37468 ms

RAG

Too heavy

5.1 tok/s

69145 ms

Quantization options

How Granite Code 20B (20B params) fits at each quantization level on MacBook Pro M3 Pro 18GB (13.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_KBest for your GPU	2	7.8 GB	Low	A81
Q3_K_S	3	9.8 GB	Low	F0
NVFP4	4	11.2 GB	Medium	F0
Q4_K_M	4	12.2 GB	Medium	F0
Q5_K_M	5	14.4 GB	High	F0
Q6_K	6	16.4 GB	High	F0
Q8_0	8	21.4 GB	Very High	F0
F16	16	41.0 GB	Maximum	F0