Arao AI
Back to search results

Qwen3.5-27B · IQ3_S

Start a discussion
NVIDIA DGX Spark128GBOllama v0.36L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.5-27B (IQ3_S) on NVIDIA DGX Spark with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

119.95 tok/s

Decode

Decode speed

607.06 tok/s

Prefill

Prefill speed

2.23 s

TTFT

Time to first token

39.52 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

2 days ago

Last verified

Performance

  1. Ollama · IQ3_S (current)Decode 119.95 · Prefill 607.06 ·
  2. Strata · Q4_K_S Decode 169.69 · Prefill 928.54 ·

Core figures

Decode (avg)
119.95 tok/s
Prefill (avg)
607.06 tok/s
TTFT (avg)
2.23 s
VRAM (avg)
39.52 GB
MTP acceptance rate
—
TTFB
0.96 GB
Power draw
206 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.5-27B
Quantization
IQ3_S
Framework
Ollama
Version
v0.36
Context length
4096 tokens
Batch size
256
GPU layers
99
Flash Attention
Off

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)
39.52 GB
OS
Windows 11 23H2
Driver
555.85
CUDA
12.6
Power draw
206 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-16

    v0.36 · Windows 11 23H2 · CUDA 12.6 · 4096 ctx

    119.95 tok/s

    Decode

    607.06 tok/s

    Prefill

    2.23 s

    TTFT

    39.52 GB

    VRAM

    —

    MTP

    206 W

    Power draw

    L2 batch seed 2026-10-16: Ollama on hw_id=54 running model_id=50 at IQ3_S