Arao AI
Back to search results

Qwen3-8B · Q3_K_M

Start a discussion
AMD Ryzen AI Max+ 395128GBOllama v0.27L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3-8B (Q3_K_M) on AMD Ryzen AI Max+ 395 with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

34.18 tok/s

Decode

Decode speed

132.79 tok/s

Prefill

Prefill speed

0.58 s

TTFT

Time to first token

13.52 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

17 days ago

Last verified

Performance

  1. vLLM · Q3_K_S Decode 38.79 · Prefill 186.13 ·
  2. Ollama · FP16 Decode 14.12 · Prefill 69.25 ·
  3. Ollama · Q3_K_M (current)Decode 34.18 · Prefill 132.79 ·
  4. llama.cpp · FP16 Decode 52.16 · Prefill 223.09 ·
  5. llama.cpp · Q8_0 Decode 70.95 · Prefill 365.06 ·
  6. vLLM · IQ3_S Decode 125.05 · Prefill 672.6 ·
  7. Ollama · IQ3_S Decode 106.51 · Prefill 487.62 ·

Core figures

Decode (avg)
34.18 tok/s
Prefill (avg)
132.79 tok/s
TTFT (avg)
0.58 s
VRAM (avg)
13.52 GB
MTP acceptance rate
—
TTFB
0.27 GB
Power draw
48.6 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3-8B
Quantization
Q3_K_M
Framework
Ollama
Version
v0.27
Context length
32768 tokens
Batch size
512
GPU layers
33
Flash Attention
On

Hardware

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

GPU
AMD Ryzen AI Max+ 395
Nominal VRAM
128 GB
Measured VRAM (avg)
13.52 GB
OS
Windows 11 23H2
Driver
amdgpu-pro 24.20
CUDA
12.4
Power draw
48.6 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-19

    v0.27 · Windows 11 23H2 · CUDA 12.4 · 32768 ctx

    34.18 tok/s

    Decode

    132.79 tok/s

    Prefill

    0.58 s

    TTFT

    13.52 GB

    VRAM

    —

    MTP

    48.6 W

    Power draw

    L2 batch seed 2026-10-19: Ollama on hw_id=24 running model_id=53 at Q3_K_M