Can GPT-OSS 20B run on NVIDIA DGX Spark 128GB?

YES — With F16

S86Excellent

Estimated from fit model

GPT-OSS 20B needs ~59.7 GB VRAM. NVIDIA DGX Spark 128GB has 0 MB. With F16 quantization, expect ~13 tok/s.

Runtime: OllamaCapacity: RoomyBandwidth: LowStack: BasicBottleneck: 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.

GPT-OSS 20B at Q4_K_M needs 16.5 GB — too much for NVIDIA DGX Spark 128GB (0.0 GB). Runs at F16 (59.7 GB) with maximum quality. 8 quantization levels fit.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 29.5 GB, 31.4 tok/s, Runs well

29.5 GB required108.8 GB available

27% VRAM used

Fit status

Runs well

Decode

31.4 tok/s

TTFT

6157 ms

Safe context

128K

Memory

29.5 GB / 108.8 GB

Memory breakdown

Weights12.8 GB

KV Cache2.4 GB

Runtime1.2 GB

Headroom13.1 GB

See how fast it feels

See how fast it feelsGPT-OSS 20B 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: 31.4 tok/s decode · 6.2s TTFT (warm) · 79 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	A	Runs well	31.4 tok/s	3358 ms	128K
Coding	F	Too heavy	5.3 tok/s	36770 ms	4K
Agentic Coding	A	Runs well	31.4 tok/s	8955 ms	128K
Reasoning	A	Runs well	31.4 tok/s	7276 ms	128K
RAG	A	Runs well	31.4 tok/s	11194 ms	128K

Quantization options

How GPT-OSS 20B (21B params) fits at each quantization level on NVIDIA DGX Spark 128GB (92.2 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	8.2 GB	Low	A78
Q3_K_S	3	10.3 GB	Low	A78
NVFP4	4	11.8 GB	Medium	A78
Q4_K_M	4	12.8 GB	Medium	A78
Q5_K_M	5	15.1 GB	High	A79
Q6_K	6	17.2 GB	High	A79
Q8_0	8	22.5 GB	Very High	A80
F16Best for your GPU	16	43.1 GB	Maximum	A84

Get started

Copy-paste commands to run GPT-OSS 20B on your machine.

Run

ollama run gpt-oss

Opciones de mejora

Hardware que ejecuta bien GPT-OSS 20B

NVIDIA H200 141GBOpción económica

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

Sube la velocidad estimada de decodificación alrededor de un 2365%.774 tok/s decodificación

Sube la velocidad estimada de decodificación alrededor de un 2365%.

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 2365%.774 tok/s decodificación

Sube la velocidad estimada de decodificación alrededor de un 2365%.

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 4008%.1290 tok/s decodificación

Sube la velocidad estimada de decodificación alrededor de un 4008%.

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 GPT-OSS 20B?

Yes, NVIDIA DGX Spark 128GB can run GPT-OSS 20B at F16 quantization (Runs well). The recommended Q4_K_M requires 16.5 GB which exceeds available memory, but at F16 it needs only 59.7 GB. Expected decode speed: 13.1 tok/s.

How much VRAM does GPT-OSS 20B need?

GPT-OSS 20B (21B parameters) requires approximately 16.5 GB at Q4_K_M quantization. On NVIDIA DGX Spark 128GB, it fits at F16 using 59.7 GB.

What is the best quantization for GPT-OSS 20B?

The recommended quantization is Q4_K_M, but on NVIDIA DGX Spark 128GB the best fitting quantization is F16, which uses 59.7 GB.

What speed will GPT-OSS 20B run at on NVIDIA DGX Spark 128GB?

On NVIDIA DGX Spark 128GB, GPT-OSS 20B achieves approximately 13.1 tokens per second decode speed with a time-to-first-token of 14779ms using F16 quantization.

Can NVIDIA DGX Spark 128GB run GPT-OSS 20B for coding?

For coding workloads, GPT-OSS 20B on NVIDIA DGX Spark 128GB receives a F grade with 5.3 tok/s and 4K context.

What context window can GPT-OSS 20B use on NVIDIA DGX Spark 128GB?

On NVIDIA DGX Spark 128GB, GPT-OSS 20B can safely use up to 128K tokens of context at F16 quantization. The model's official context limit is 128K, but available memory constrains the safe maximum.

Is unified memory on NVIDIA DGX Spark 128GB as fast as VRAM for GPT-OSS 20B?

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 GPT-OSS 20B

Embed this result▼

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

<iframe src="https://willitrunai.com/embed/gpt-oss-20b-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: