Can GPT-OSS 120B run on NVIDIA A40 48GB?

YES — With Q2_K

A76Great

○Estimated from fit model

GPT-OSS 120B needs ~56.2 GB VRAM. NVIDIA A40 48GB has 48.0 GB. With Q2_K quantization, expect ~6 tok/s.

Runtime: llama.cppCapacity: OffloadBandwidth: MediumStack: 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.

GPT-OSS 120B at Q4_K_M needs 82.0 GB — too much for NVIDIA A40 48GB (48.0 GB). Runs at Q2_K (56.2 GB) with low quality.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 82.0 GB, exceeds 48.0 GB available

82.0 GB required48.0 GB available

171% VRAM needed

34.0 GB over capacity — needs offload or smaller quantization

Fit status

Too heavy

Decode

2.0 tok/s

TTFT

96220 ms

Safe context

Memory

82.0 GB / 48.0 GB

Offload

40%

Memory breakdown

Weights71.4 GB

KV Cache4.9 GB

Runtime0.9 GB

Headroom4.8 GB

See how fast it feels

With memory offload — actual speed may be lower

See how fast it feelsGPT-OSS 120B on NVIDIA A40 48GB
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: 2.0 tok/s decode · 96.2s TTFT (warm) · 5 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.

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 6.7 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	2.1 tok/s	49247 ms	4K
Coding	F	Too heavy	2.0 tok/s	96220 ms	4K
Agentic Coding	F	Too heavy	2.0 tok/s	140800 ms	4K
Reasoning	F	Too heavy	2.0 tok/s	113715 ms	4K
RAG	F	Too heavy	2.0 tok/s	176000 ms	4K

Quantization options

How GPT-OSS 120B (117B params) fits at each quantization level on NVIDIA A40 48GB (48.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	45.6 GB	Low	F0
Q3_K_S	3	57.3 GB	Low	F0
NVFP4	4	65.5 GB	Medium	F0
Q4_K_M	4	71.4 GB	Medium	F0
Q5_K_M	5	84.2 GB	High	F0
Q6_K	6	95.9 GB	High	F0
Q8_0	8	125.2 GB	Very High	F0
F16	16	239.8 GB	Maximum	F0

Get started

Copy-paste commands to run GPT-OSS 120B on your machine.

Run

ollama run gpt-oss:120b

Upgrade options

Hardware that runs GPT-OSS 120B well

👁 NVIDIA
RTX PRO 6000 Blackwell Workstation Edition 96GBBudget pick

96 GB VRAM (+48)1792 GB/s (+1096)

Makes the model fit on the accelerator instead of staying completely out of reach.22.9 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.

~$9,999 MSRP

👁 NVIDIA
RTX PRO 6000 Blackwell Server Edition 96GBBest value

96 GB VRAM (+48)1597 GB/s (+901)

Makes the model fit on the accelerator instead of staying completely out of reach.20.4 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.

~$9,999 MSRP

👁 NVIDIA
NVIDIA H20 96GBNVIDIA upgrade

96 GB VRAM (+48)4000 GB/s (+3304)

Makes the model fit on the accelerator instead of staying completely out of reach.49.4 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

See all results for NVIDIA A40 48GB See all hardware for GPT-OSS 120B

URL: https://willitrunai.com/can-run/gpt-oss-120b-on-a40-48gb

⇱ GPT-OSS 120B on NVIDIA A40 48GB? No — Alternatives

Can GPT-OSS 120B run on NVIDIA A40 48GB?

YES — With Q2_K

Choose the run profile you care about

Memory breakdown

See how fast it feels

What limits this setup

Best improvement path

Performance by workload

Quantization options

Get started

Hardware that runs GPT-OSS 120B well

Frequently asked questions