Arao AI
Back to search results

Qwen3.8-9B · Q8_0

Start a discussion
NVIDIA DGX Spark128GBOllama v0.38L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.8-9B (Q8_0) on NVIDIA DGX Spark with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

88.03 tok/s

Decode

Decode speed

479.67 tok/s

Prefill

Prefill speed

1.19 s

TTFT

Time to first token

23.84 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

5 days ago

Last verified

Performance

  1. vLLM · IQ3_S Decode 178.22 · Prefill 972.1 ·
  2. vLLM · Q5_K_M Decode 122.43 · Prefill 399.31 ·
  3. Strata · Q3_K_S Decode 245.76 · Prefill 803.01 ·
  4. Ollama · Q8_0 (current)Decode 88.03 · Prefill 479.67 ·

Core figures

Decode (avg)
88.03 tok/s
Prefill (avg)
479.67 tok/s
TTFT (avg)
1.19 s
VRAM (avg)
23.84 GB
MTP acceptance rate
—
TTFB
0.7 GB
Power draw
266.4 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.8-9B
Quantization
Q8_0
Framework
Ollama
Version
v0.38
Context length
8192 tokens
Batch size
512
GPU layers
99
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)
23.84 GB
OS
Ubuntu 24.04
Driver
565.57
CUDA
12.8
Power draw
266.4 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

    v0.38 · Ubuntu 24.04 · CUDA 12.8 · 8192 ctx

    88.03 tok/s

    Decode

    479.67 tok/s

    Prefill

    1.19 s

    TTFT

    23.84 GB

    VRAM

    —

    MTP

    266.4 W

    Power draw

    L2 batch seed 2026-10-13: Ollama on hw_id=54 running model_id=52 at Q8_0