Can Codestral 22B v0.1 run on Mac Studio M2 Ultra 128GB?

YES — Runs Great

C46Usable

Estimated from fit model

Codestral 22B v0.1 needs ~30.7 GB VRAM. Mac Studio M2 Ultra 128GB has 92.2 GB. With Q4_K_M quantization, expect ~35 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) — 30.7 GB, 34.6 tok/s, Runs well

30.7 GB required92.2 GB available

33% VRAM used

Fit status

Runs well

Decode

34.6 tok/s

TTFT

5599 ms

Safe context

397K

Memory

30.7 GB / 92.2 GB

Memory breakdown

Weights13.4 GB

KV Cache2.6 GB

Runtime0.9 GB

Headroom13.8 GB

See how fast it feels

See how fast it feelsCodestral 22B v0.1 on Mac Studio M2 Ultra 128GB
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: 34.6 tok/s decode · 5.6s TTFT (warm) · 86 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	34.6 tok/s	3054 ms	397K
Coding	C	Runs well	34.6 tok/s	5599 ms	397K
Agentic Coding	C	Runs well	34.6 tok/s	8145 ms	397K
Reasoning	C	Runs well	34.6 tok/s	6617 ms	397K
RAG	C	Runs well	34.6 tok/s	10181 ms	397K

Inference speed

Codestral 22B v0.1 inference speed — tokens per second by GPU & Mac

Estimated decode speed (tokens/sec) for Codestral 22B v0.1 at Q4_K_M across popular GPUs and Apple Silicon, using the fastest local runtime per device. Fastest is RTX 5090 32GB at ~90 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	89.5	Fits
RTX 4090 24GB	24 GB	Q4_K_M	57.1	Fits
RX 7900 XTX 24GB	24 GB	Q4_K_M	51.5	Fits
RTX 3090 24GB	24 GB	Q4_K_M	48.8	Fits
Mac Studio M3 Ultra 256GB	256 GB	Q4_K_M	41.5	Fits
MacBook Pro M4 Max 128GB	128 GB	Q4_K_M	34.8	Fits
MacBook Pro M4 Max 64GB	64 GB	Q4_K_M	34.8	Fits
Mac Studio M2 Ultra 128GB	128 GB	Q4_K_M	34.6	Fits
Mac Studio M1 Ultra 128GB	128 GB	Q4_K_M	32.8	Fits
RTX 4080 Super 16GB	16 GB	Q4_K_M	24.3	Heavy offload
MacBook Pro M4 Pro 48GB	48 GB	Q4_K_M	21.9	Fits
MacBook Pro M3 Max 64GB	64 GB	Q4_K_M	17.9	Fits
MacBook Pro M1 Max 64GB	64 GB	Q4_K_M	16.4	Fits
RTX 4070 12GB	12 GB	Q4_K_M	8.6	Too big
RTX 3060 12GB	12 GB	Q4_K_M	5.4	Too big
RTX 4060 8GB	8 GB	Q4_K_M	2.2	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 Codestral 22B v0.1 (22B params) fits at each quantization level on Mac Studio M2 Ultra 128GB (92.2 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	8.6 GB	Low	D39
Q3_K_S	3	10.8 GB	Low	D39
NVFP4	4	12.3 GB	Medium	D40
Q4_K_M	4	13.4 GB	Medium	D40
Q5_K_M	5	15.8 GB	High	C40
Q6_K	6	18.0 GB	High	C40
Q8_0	8	23.5 GB	Very High	C41
F16Best for your GPU	16	45.1 GB	Maximum	C46

Get started

Copy-paste commands to run Codestral 22B v0.1 on your machine.

Run

lms load hf-sanctumai--codestral-22b-v0-1-gguf && lms server start

Opções de upgrade

Hardware que roda bem Codestral 22B v0.1

RTX PRO 6000 Blackwell Workstation Edition 96GBOpção econômica

1792 GB/s (+992)

Raises estimated decode speed by about 224%.112.2 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%.100 tok/s decodificação

Raises estimated decode speed by about 189%.

~$9,999 MSRP

Frequently asked questions

Can Mac Studio M2 Ultra 128GB run Codestral 22B v0.1?

Yes, Mac Studio M2 Ultra 128GB can run Codestral 22B v0.1 with a C grade (Runs well). Expected decode speed: 34.6 tok/s.

How much VRAM does Codestral 22B v0.1 need?

Codestral 22B v0.1 (22B parameters) requires approximately 30.7 GB of memory with Q4_K_M quantization.

What is the best quantization for Codestral 22B v0.1?

The recommended quantization for Codestral 22B v0.1 is Q4_K_M, which balances quality and memory efficiency.

What speed will Codestral 22B v0.1 run at on Mac Studio M2 Ultra 128GB?

On Mac Studio M2 Ultra 128GB, Codestral 22B v0.1 achieves approximately 34.6 tokens per second decode speed with a time-to-first-token of 5599ms using Q4_K_M quantization.

Can Mac Studio M2 Ultra 128GB run Codestral 22B v0.1 for coding?

For coding workloads, Codestral 22B v0.1 on Mac Studio M2 Ultra 128GB receives a C grade with 34.6 tok/s and 397K context.

What context window can Codestral 22B v0.1 use on Mac Studio M2 Ultra 128GB?

On Mac Studio M2 Ultra 128GB, Codestral 22B v0.1 can safely use up to 397K 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 Codestral 22B v0.1?

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 Codestral 22B v0.1

Embed this result▼

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

<iframe src="https://willitrunai.com/embed/hf-sanctumai--codestral-22b-v0-1-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: