Arao AI
Back to search results

Qwen3.5-27B · Q4_K_S

Start a discussion
NVIDIA DGX Spark128GBStrata v0.33L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.5-27B (Q4_K_S) on NVIDIA DGX Spark with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.

169.69 tok/s

Decode

Decode speed

928.54 tok/s

Prefill

Prefill speed

1.84 s

TTFT

Time to first token

46.33 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 Decode 119.95 · Prefill 607.06 ·
  2. Strata · Q4_K_S (current)Decode 169.69 · Prefill 928.54 ·

Core figures

Decode (avg)
169.69 tok/s
Prefill (avg)
928.54 tok/s
TTFT (avg)
1.84 s
VRAM (avg)
46.33 GB
MTP acceptance rate
0.71
TTFB
1.36 GB
Power draw
183.7 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.5-27B
Quantization
Q4_K_S
Framework
Strata
Version
v0.33
Context length
4096 tokens
Batch size
256
GPU layers
15
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)
46.33 GB
OS
Ubuntu 24.04
Driver
570.86
CUDA
12.8
Power draw
183.7 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

    v0.33 · Ubuntu 24.04 · CUDA 12.8 · 4096 ctx

    169.69 tok/s

    Decode

    928.54 tok/s

    Prefill

    1.84 s

    TTFT

    46.33 GB

    VRAM

    0.71

    MTP

    183.7 W

    Power draw

    L2 batch seed 2026-10-16: Strata on hw_id=54 running model_id=50 at Q4_K_S