Can GPT-OSS 120B run on AMD Instinct MI210 64GB?

YES — With Q3_K_S

A78Great

Estimated from fit model

GPT-OSS 120B needs ~69.5 GB VRAM. AMD Instinct MI210 64GB has 64.0 GB. With Q3_K_S quantization, expect ~12 tok/s.

Runtime: llama.cppCapacity: OffloadBandwidth: HighStack: 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 83.6 GB — too much for AMD Instinct MI210 64GB (64.0 GB). Runs at Q3_K_S (69.5 GB) with low quality. 2 quantization levels fit.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 83.6 GB, exceeds 64.0 GB available

83.6 GB required64.0 GB available

131% VRAM needed

19.6 GB over capacity — needs offload or smaller quantization

Fit status

Too heavy

Decode

7.3 tok/s

TTFT

26658 ms

Safe context

Memory

83.6 GB / 64.0 GB

Offload

20%

Memory breakdown

Weights71.4 GB

KV Cache4.9 GB

Runtime0.9 GB

Headroom6.4 GB

See how fast it feels

With memory offload — actual speed may be lower

See how fast it feelsGPT-OSS 120B on AMD Instinct MI210 64GB
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: 7.3 tok/s decode · 26.7s TTFT (warm) · 18 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 4.5 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	7.7 tok/s	13661 ms	4K
Coding	F	Too heavy	7.3 tok/s	26658 ms	4K
Agentic Coding	F	Too heavy	6.4 tok/s	43700 ms	4K
Reasoning	F	Too heavy	7.3 tok/s	31505 ms	4K
RAG	F	Too heavy	6.4 tok/s	54624 ms	4K

Quantization options

How GPT-OSS 120B (117B params) fits at each quantization level on AMD Instinct MI210 64GB (64.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_KBest for your GPU	2	45.6 GB	Low	S88
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-Optionen

Hardware, die GPT-OSS 120B gut ausführt

AMD Instinct MI350X 288GBGünstige Wahl

288 GB VRAM (+224)8000 GB/s (+6362)

Makes the model fit on the accelerator instead of staying completely out of reach.89 tok/s Dekodierung

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.

ca. $8,000 MSRP

AMD Instinct MI300A 128GBBestes Preis-Leistungs-Verhältnis

128 GB VRAM (+64)5300 GB/s (+3662)

Makes the model fit on the accelerator instead of staying completely out of reach.56.5 tok/s Dekodierung

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.

ca. $12,000 MSRP

AMD Instinct MI250X 128GBAMD-Upgrade

128 GB VRAM (+64)3200 GB/s (+1562)

Makes the model fit on the accelerator instead of staying completely out of reach.38 tok/s Dekodierung

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.

ca. $15,000 MSRP

Frequently asked questions

Can AMD Instinct MI210 64GB run GPT-OSS 120B?

Yes, AMD Instinct MI210 64GB can run GPT-OSS 120B at Q3_K_S quantization (Very compromised (needs ~4.5 GB host RAM)). The recommended Q4_K_M requires 83.6 GB which exceeds available memory, but at Q3_K_S it needs only 69.5 GB. Expected decode speed: 12.4 tok/s.

How much VRAM does GPT-OSS 120B need?

GPT-OSS 120B (117B parameters) requires approximately 83.6 GB at Q4_K_M quantization. On AMD Instinct MI210 64GB, it fits at Q3_K_S using 69.5 GB.

What is the best quantization for GPT-OSS 120B?

The recommended quantization is Q4_K_M, but on AMD Instinct MI210 64GB the best fitting quantization is Q3_K_S, which uses 69.5 GB.

What speed will GPT-OSS 120B run at on AMD Instinct MI210 64GB?

On AMD Instinct MI210 64GB, GPT-OSS 120B achieves approximately 12.4 tokens per second decode speed with a time-to-first-token of 15633ms using Q3_K_S quantization.

Can AMD Instinct MI210 64GB run GPT-OSS 120B for coding?

For coding workloads, GPT-OSS 120B on AMD Instinct MI210 64GB receives a F grade with 7.3 tok/s and 4K context.

What context window can GPT-OSS 120B use on AMD Instinct MI210 64GB?

On AMD Instinct MI210 64GB, GPT-OSS 120B can safely use up to 4K tokens of context at Q3_K_S quantization. The model's official context limit is 131K, but available memory constrains the safe maximum.

What should I upgrade first if GPT-OSS 120B feels slow on AMD Instinct MI210 64GB?

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.

See all results for AMD Instinct MI210 64GB See all hardware for GPT-OSS 120B

Embed this result▼

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

<iframe src="https://willitrunai.com/embed/gpt-oss-120b-on-instinct-mi210-64gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview: