Arao AI
Back to search results

Qwen3-4B · IQ1_S

Start a discussion
NVIDIA DGX Spark128GBStrata v0.46L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3-4B (IQ1_S) on NVIDIA DGX Spark with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.

323.83 tok/s

Decode

Decode speed

1,439.4 tok/s

Prefill

Prefill speed

0.86 s

TTFT

Time to first token

5.63 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

5 days ago

Last verified

Performance

  1. llama.cpp · Q3_K_M Decode 206.5 · Prefill 1076.24 ·
  2. Strata · Q3_K_S Decode 338.04 · Prefill 1544.38 ·
  3. Strata · Q5_K_M Decode 221.01 · Prefill 864.66 ·
  4. vLLM · Q3_K_L Decode 202.71 · Prefill 852.86 ·
  5. Strata · IQ1_S (current)Decode 323.83 · Prefill 1439.4 ·

Core figures

Decode (avg)
323.83 tok/s
Prefill (avg)
1,439.4 tok/s
TTFT (avg)
0.86 s
VRAM (avg)
5.63 GB
MTP acceptance rate
0.49
TTFB
2.59 GB
Power draw
200.4 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3-4B
Quantization
IQ1_S
Framework
Strata
Version
v0.46
Context length
32768 tokens
Batch size
1024
GPU layers
5
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)
5.63 GB
OS
Ubuntu 24.04
Driver
555.85
CUDA
12.8
Power draw
200.4 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

    v0.46 · Ubuntu 24.04 · CUDA 12.8 · 32768 ctx

    323.83 tok/s

    Decode

    1,439.4 tok/s

    Prefill

    0.86 s

    TTFT

    5.63 GB

    VRAM

    0.49

    MTP

    200.4 W

    Power draw

    L2 batch seed 2026-10-13: Strata on hw_id=54 running model_id=54 at IQ1_S