Arao AI
Back to search results

Qwen3.8-9B · IQ4_XS

Start a discussion
NVIDIA RTX 509032GBOllama v0.50L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.8-9B (IQ4_XS) on NVIDIA RTX 5090 with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

224.73 tok/s

Decode

Decode speed

806.05 tok/s

Prefill

Prefill speed

2.14 s

TTFT

Time to first token

16.09 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. Ollama · IQ4_XS (current)Decode 224.73 · Prefill 806.05 ·
  2. Ollama · Q5_K_S Decode 182.89 · Prefill 696.63 ·
  3. Strata · IQ1_S Decode 329.29 · Prefill 1427.88 ·
  4. Strata · Q5_K_S Decode 276.86 · Prefill 910.66 ·

Core figures

Decode (avg)
224.73 tok/s
Prefill (avg)
806.05 tok/s
TTFT (avg)
2.14 s
VRAM (avg)
16.09 GB
MTP acceptance rate
—
TTFB
1.8 GB
Power draw
565.4 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.8-9B
Quantization
IQ4_XS
Framework
Ollama
Version
v0.50
Context length
2048 tokens
Batch size
256
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 RTX 5090
Nominal VRAM
32 GB
Measured VRAM (avg)
16.09 GB
OS
macOS 15.0
Driver
565.57
CUDA
12.8
Power draw
565.4 W

Sources & evidence

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

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

    v0.50 · macOS 15.0 · CUDA 12.8 · 2048 ctx

    224.73 tok/s

    Decode

    806.05 tok/s

    Prefill

    2.14 s

    TTFT

    16.09 GB

    VRAM

    —

    MTP

    565.4 W

    Power draw

    L2 batch seed 2026-10-10: Ollama on hw_id=13 running model_id=52 at IQ4_XS