Arao AI
Back to search results

Qwen3.8-9B · Q3_K_M

Start a discussion
AMD Strix Point (Radeon 890M)vLLM v0.36L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.8-9B (Q3_K_M) on AMD Strix Point (Radeon 890M) with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

62.84 tok/s

Decode

Decode speed

312.12 tok/s

Prefill

Prefill speed

0.95 s

TTFT

Time to first token

14.95 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

2 days ago

Last verified

Performance

  1. vLLM · Q2_K Decode 74.53 · Prefill 292.59 ·
  2. vLLM · Q3_K_M (current)Decode 62.84 · Prefill 312.12 ·
  3. Strata · Q4_K_M Decode 68.09 · Prefill 270.12 ·

Core figures

Decode (avg)
62.84 tok/s
Prefill (avg)
312.12 tok/s
TTFT (avg)
0.95 s
VRAM (avg)
14.95 GB
MTP acceptance rate
—
TTFB
0.5 GB
Power draw
62.2 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.8-9B
Quantization
Q3_K_M
Framework
vLLM
Version
v0.36
Context length
8192 tokens
Batch size
8
GPU layers
999
Flash Attention
On

Hardware

Nominal and measured figures are shown side by side; whether it runs is the reader's call

GPU
AMD Strix Point (Radeon 890M)
Measured VRAM (avg)
14.95 GB
OS
macOS 15.0
Driver
amdgpu 6.10
CUDA
N/A
Power draw
62.2 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 · macOS 15.0 · CUDA N/A · 8192 ctx

    62.84 tok/s

    Decode

    312.12 tok/s

    Prefill

    0.95 s

    TTFT

    14.95 GB

    VRAM

    —

    MTP

    62.2 W

    Power draw

    L2 batch seed 2026-10-16: vLLM on hw_id=89 running model_id=52 at Q3_K_M