Arao AI
Back to search results

gemma-4-12B-it · Q2_K

Start a discussion
NVIDIA DGX Spark128GBOllama v0.28L2 Cross-framework verified

This page aggregates 1 real-world runs of gemma-4-12B-it (Q2_K) on NVIDIA DGX Spark with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

153.5 tok/s

Decode

Decode speed

551.49 tok/s

Prefill

Prefill speed

0.59 s

TTFT

Time to first token

15.96 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

5 days ago

Last verified

Performance

  1. Ollama · Q2_K (current)Decode 153.5 · Prefill 551.49 ·
  2. Strata · Q4_K_S Decode 156.45 · Prefill 581.38 ·
  3. Strata · Q5_K_M Decode 139.24 · Prefill 652.49 ·

Core figures

Decode (avg)
153.5 tok/s
Prefill (avg)
551.49 tok/s
TTFT (avg)
0.59 s
VRAM (avg)
15.96 GB
MTP acceptance rate
—
TTFB
1.23 GB
Power draw
220.3 W

Configuration

Member-level fields are taken from the most recent run

Model
gemma-4-12B-it
Quantization
Q2_K
Framework
Ollama
Version
v0.28
Context length
32768 tokens
Batch size
256
GPU layers
28
Flash Attention
On

Hardware

Nominal and measured figures are shown side by side; whether it runs is the reader's call

GPU
NVIDIA DGX Spark
Nominal VRAM
128 GB
Measured VRAM (avg)
15.96 GB
OS
Ubuntu 24.04
Driver
565.57
CUDA
12.8
Power draw
220.3 W

Sources & evidence

1 measured records in total, each traceable to its original source

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-13

    v0.28 · Ubuntu 24.04 · CUDA 12.8 · 32768 ctx

    153.5 tok/s

    Decode

    551.49 tok/s

    Prefill

    0.59 s

    TTFT

    15.96 GB

    VRAM

    —

    MTP

    220.3 W

    Power draw

    L2 batch seed 2026-10-13: Ollama on hw_id=54 running model_id=65 at Q2_K