Can Gemma 4 26B A4B run on Gaudi 3 128GB?

YES — Runs Great

A82Great

○Estimated from fit model

Gemma 4 26B A4B needs ~32.7 GB VRAM. Gaudi 3 128GB has 128.0 GB. With Q4_K_M quantization, expect ~401 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) — 32.7 GB, 420.6 tok/s, Runs well

32.7 GB required128.0 GB available

26% VRAM used

Fit status

Runs well

Decode

420.6 tok/s

TTFT

460 ms

Safe context

256K

Memory

32.7 GB / 128.0 GB

Memory breakdown

Weights15.4 GB

KV Cache3.7 GB

Runtime0.9 GB

Headroom12.8 GB

See how fast it feels

See how fast it feelsGemma 4 26B A4B on Gaudi 3 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: 420.6 tok/s decode · 460ms TTFT (warm) · 1052 tok/s prefill

What limits this setup

The raw memory story may look fine, but the software ecosystem is still a constraint here.

Runtime ecosystem is narrower than CUDA

Intel GPUs can look attractive on memory per dollar, but local AI tooling, kernels, and model coverage are still broader and easier on CUDA today.

Best improvement path

Prefer CUDA if you want the path of least resistance

If your goal is maximum runtime coverage, easier troubleshooting, and better support for new local AI releases, CUDA is usually still the safer upgrade path.

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Chat	A	Runs well	420.6 tok/s	350 ms	256K
Coding	A	Runs well	400.6 tok/s	483 ms	256K
Agentic Coding	A	Runs well	420.6 tok/s	670 ms	256K
Reasoning	A	Runs well	420.6 tok/s	544 ms	256K
RAG	A	Runs well	420.6 tok/s	837 ms	256K

Quantization options

How Gemma 4 26B A4B (25.200000762939453B params) fits at each quantization level on Gaudi 3 128GB (128.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	9.8 GB	Low	A73
Q3_K_S	3	12.3 GB	Low	A74
NVFP4	4

Get started

Copy-paste commands to run Gemma 4 26B A4B on your machine.

Run

ollama run gemma4:26b

Your hardware

More models your Gaudi 3 128GB can run

Model	Params	Grade	Decode	Capabilities
👁 Mistral Devstral 2 123B Instruct	123B	S	37.5 tok/s
👁 Alibaba Qwen3-Coder 30B A3B Instruct	30.5B	S

Frequently asked questions

See all results for Gaudi 3 128GB See all hardware for Gemma 4 26B A4B

URL: https://willitrunai.com/can-run/gemma-4-26b-a4b-on-gaudi-3-128gb

⇱