Can Devstral 2 123B Instruct run on MacBook Pro M2 Max 96GB?

YES — With Q3_K_S

A77Great

Estimated from fit model

Devstral 2 123B Instruct needs ~76.9 GB VRAM. MacBook Pro M2 Max 96GB has 69.1 GB. With Q3_K_S quantization, expect ~3 tok/s.

Runtime: llama.cppCapacity: OffloadBandwidth: 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.

Devstral 2 123B Instruct at Q4_K_M needs 91.7 GB — too much for MacBook Pro M2 Max 96GB (69.1 GB). Runs at Q3_K_S (76.9 GB) with low quality. 2 quantization levels fit.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 91.7 GB, exceeds 69.1 GB available

91.7 GB required69.1 GB available

133% VRAM needed

22.6 GB over capacity — needs offload or smaller quantization

Fit status

Too heavy

Decode

2.3 tok/s

TTFT

85374 ms

Safe context

Memory

91.7 GB / 69.1 GB

Offload

20%

Memory breakdown

Weights75.0 GB

KV Cache5.4 GB

Runtime0.9 GB

Headroom10.4 GB

See how fast it feels

With memory offload — actual speed may be lower

See how fast it feelsDevstral 2 123B Instruct on MacBook Pro M2 Max 96GB
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: 2.3 tok/s decode · 85.4s TTFT (warm) · 6 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 6.1 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	2.3 tok/s	44946 ms	4K
Coding	F	Too heavy	2.3 tok/s	85374 ms	4K
Agentic Coding	F	Too heavy	2.1 tok/s	132742 ms	4K
Reasoning	F	Too heavy	2.3 tok/s	100896 ms	4K
RAG	F	Too heavy	2.1 tok/s	165927 ms	4K

Quantization options

How Devstral 2 123B Instruct (123B params) fits at each quantization level on MacBook Pro M2 Max 96GB (69.1 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_KBest for your GPU	2	48.0 GB	Low	S91
Q3_K_S	3	60.3 GB	Low	F0
NVFP4	4	68.9 GB	Medium	F0
Q4_K_M	4	75.0 GB	Medium	F0
Q5_K_M	5	88.6 GB	High	F0
Q6_K	6	100.9 GB	High	F0
Q8_0	8	131.6 GB	Very High	F0
F16	16	252.2 GB	Maximum	F0

Get started

Copy-paste commands to run Devstral 2 123B Instruct on your machine.

Run

lms load Devstral-2-123B-Instruct-2512 && lms server start

Upgrade options

Hardware that runs Devstral 2 123B Instruct well

MacBook Pro M3 Max 128GBBudget pick

128 GB Unified (+32)

Makes the model fit on the accelerator instead of staying completely out of reach.3.3 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.

~$2,499 MSRP

Mac Studio M2 Ultra 128GBBest value

128 GB Unified (+32)800 GB/s (+400)

Makes the model fit on the accelerator instead of staying completely out of reach.6.3 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.

~$3,999 MSRP

Mac Studio M1 Ultra 128GBApple upgrade

128 GB Unified (+32)800 GB/s (+400)

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

~$3,999 MSRP

AMD Instinct MI300A 128GBBiggest leap

128 GB VRAM (+32)5300 GB/s (+4900)

Makes the model fit on the accelerator instead of staying completely out of reach.53.8 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.

~$12,000 MSRP

Frequently asked questions

Can MacBook Pro M2 Max 96GB run Devstral 2 123B Instruct?

Yes, MacBook Pro M2 Max 96GB can run Devstral 2 123B Instruct at Q3_K_S quantization (Very compromised (needs ~6.1 GB host RAM)). The recommended Q4_K_M requires 91.7 GB which exceeds available memory, but at Q3_K_S it needs only 76.9 GB. Expected decode speed: 3.3 tok/s.

How much VRAM does Devstral 2 123B Instruct need?

Devstral 2 123B Instruct (123B parameters) requires approximately 91.7 GB at Q4_K_M quantization. On MacBook Pro M2 Max 96GB, it fits at Q3_K_S using 76.9 GB.

What is the best quantization for Devstral 2 123B Instruct?

The recommended quantization is Q4_K_M, but on MacBook Pro M2 Max 96GB the best fitting quantization is Q3_K_S, which uses 76.9 GB.

What speed will Devstral 2 123B Instruct run at on MacBook Pro M2 Max 96GB?

On MacBook Pro M2 Max 96GB, Devstral 2 123B Instruct achieves approximately 3.3 tokens per second decode speed with a time-to-first-token of 59208ms using Q3_K_S quantization.

Can MacBook Pro M2 Max 96GB run Devstral 2 123B Instruct for coding?

For coding workloads, Devstral 2 123B Instruct on MacBook Pro M2 Max 96GB receives a F grade with 2.3 tok/s and 4K context.

What context window can Devstral 2 123B Instruct use on MacBook Pro M2 Max 96GB?

On MacBook Pro M2 Max 96GB, Devstral 2 123B Instruct can safely use up to 4K tokens of context at Q3_K_S quantization. The model's official context limit is 256K, but available memory constrains the safe maximum.

What should I upgrade first if Devstral 2 123B Instruct feels slow on MacBook Pro M2 Max 96GB?

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 M2 Max 96GB as fast as VRAM for Devstral 2 123B Instruct?

Not always. MacBook Pro M2 Max 96GB 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 M2 Max 96GB See all hardware for Devstral 2 123B Instruct

Embed this result▼

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

<iframe src="https://willitrunai.com/embed/devstral-2-123b-on-m2-max-96gb" 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 6.1 GB of extra host RAM just for the offloaded portion, before OS and other tools.

Workload

Grade

Fit

Decode

TTFT

Context

Chat

Too heavy

2.3 tok/s

44946 ms

Coding

Too heavy

2.3 tok/s

85374 ms

Agentic Coding

Too heavy

2.1 tok/s

132742 ms

Reasoning

Too heavy

2.3 tok/s

100896 ms

RAG

Too heavy

2.1 tok/s

165927 ms

Quantization options

How Devstral 2 123B Instruct (123B params) fits at each quantization level on MacBook Pro M2 Max 96GB (69.1 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_KBest for your GPU	2	48.0 GB	Low	S91
Q3_K_S	3	60.3 GB	Low	F0
NVFP4	4	68.9 GB	Medium	F0
Q4_K_M	4	75.0 GB	Medium	F0
Q5_K_M	5	88.6 GB	High	F0
Q6_K	6	100.9 GB	High	F0
Q8_0	8	131.6 GB	Very High	F0
F16	16	252.2 GB	Maximum	F0