Can Devstral 2 123B Instruct run on NVIDIA DGX Spark 128GB?

YES — With Q6_K

A76Great

Estimated from fit model

Devstral 2 123B Instruct needs ~120.2 GB VRAM. NVIDIA DGX Spark 128GB has 0 MB. With Q6_K quantization, expect ~2 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 81.3 GB — too much for NVIDIA DGX Spark 128GB (0.0 GB). Runs at Q6_K (120.2 GB) with high quality. 6 quantization levels fit.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 94.4 GB, 2.4 tok/s, Tight fit

94.4 GB required108.8 GB available

87% VRAM used

Fit status

Tight fit

Decode

2.4 tok/s

TTFT

81545 ms

Safe context

59K

Memory

94.4 GB / 108.8 GB

Memory breakdown

Weights75.0 GB

KV Cache5.4 GB

Runtime0.9 GB

Headroom13.1 GB

See how fast it feels

See how fast it feelsDevstral 2 123B Instruct on NVIDIA DGX Spark 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: 2.4 tok/s decode · 81.5s 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 9.6 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	S	Tight fit	2.4 tok/s	44479 ms	59K
Coding	F	Too heavy	2.0 tok/s	96800 ms	4K
Agentic Coding	S	Tight fit	2.4 tok/s	118611 ms	59K
Reasoning	S	Tight fit	2.4 tok/s	96371 ms	59K
RAG	S	Tight fit	2.4 tok/s	148264 ms	59K

Quantization options

How Devstral 2 123B Instruct (123B params) fits at each quantization level on NVIDIA DGX Spark 128GB (92.2 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	48.0 GB	Low	S91
Q3_K_S	3	60.3 GB	Low	S91
NVFP4Best for your GPU	4	68.9 GB	Medium	S91
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

Opciones de mejora

Hardware que ejecuta bien Devstral 2 123B Instruct

NVIDIA H200 141GBOpción económica

141 GB VRAM (+13)4800 GB/s (+4527)

Sube la velocidad estimada de decodificación alrededor de un 2333%.58.4 tok/s decodificación

Sube la velocidad estimada de decodificación alrededor de un 2333%.

Añade margen de memoria para más contexto y para que el modelo envejezca mejor.

~$30,000 MSRP

NVIDIA H200 PCIe 141GBMejor relación calidad-precio

141 GB VRAM (+13)4800 GB/s (+4527)

Sube la velocidad estimada de decodificación alrededor de un 2333%.58.4 tok/s decodificación

Sube la velocidad estimada de decodificación alrededor de un 2333%.

Añade margen de memoria para más contexto y para que el modelo envejezca mejor.

~$30,000 MSRP

NVIDIA B200 180GBMejora NVIDIA

180 GB VRAM (+52)8000 GB/s (+7727)

Sube la velocidad estimada de decodificación alrededor de un 3958%.97.4 tok/s decodificación

Sube la velocidad estimada de decodificación alrededor de un 3958%.

Añade margen de memoria para más contexto y para que el modelo envejezca mejor.

~$30,000 MSRP

Frequently asked questions

Can NVIDIA DGX Spark 128GB run Devstral 2 123B Instruct?

Yes, NVIDIA DGX Spark 128GB can run Devstral 2 123B Instruct at Q6_K quantization (Very compromised (needs ~9.6 GB host RAM)). The recommended Q4_K_M requires 81.3 GB which exceeds available memory, but at Q6_K it needs only 120.2 GB. Expected decode speed: 2.0 tok/s.

How much VRAM does Devstral 2 123B Instruct need?

Devstral 2 123B Instruct (123B parameters) requires approximately 81.3 GB at Q4_K_M quantization. On NVIDIA DGX Spark 128GB, it fits at Q6_K using 120.2 GB.

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

The recommended quantization is Q4_K_M, but on NVIDIA DGX Spark 128GB the best fitting quantization is Q6_K, which uses 120.2 GB.

What speed will Devstral 2 123B Instruct run at on NVIDIA DGX Spark 128GB?

On NVIDIA DGX Spark 128GB, Devstral 2 123B Instruct achieves approximately 2.0 tokens per second decode speed with a time-to-first-token of 96800ms using Q6_K quantization.

Can NVIDIA DGX Spark 128GB run Devstral 2 123B Instruct for coding?

For coding workloads, Devstral 2 123B Instruct on NVIDIA DGX Spark 128GB receives a F grade with 2.0 tok/s and 4K context.

What context window can Devstral 2 123B Instruct use on NVIDIA DGX Spark 128GB?

On NVIDIA DGX Spark 128GB, Devstral 2 123B Instruct can safely use up to 4K tokens of context at Q6_K 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 NVIDIA DGX Spark 128GB?

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 NVIDIA DGX Spark 128GB as fast as VRAM for Devstral 2 123B Instruct?

Not always. NVIDIA DGX Spark 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 NVIDIA DGX Spark 128GB 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-dgx-spark-128gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview: