Arao AI
Back to search results

Qwen3-4B-Thinking-2507 · Q4_K_S

Start a discussion
AMD Radeon AI PRO R970032GBvLLM v0.49L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3-4B-Thinking-2507 (Q4_K_S) on AMD Radeon AI PRO R9700 with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

158.57 tok/s

Decode

Decode speed

583.28 tok/s

Prefill

Prefill speed

0.64 s

TTFT

Time to first token

8.65 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

5 days ago

Last verified

Performance

  1. vLLM · Q4_K_S (current)Decode 158.57 · Prefill 583.28 ·
  2. vLLM · Q5_K_M Decode 111.48 · Prefill 571.17 ·

Core figures

Decode (avg)
158.57 tok/s
Prefill (avg)
583.28 tok/s
TTFT (avg)
0.64 s
VRAM (avg)
8.65 GB
MTP acceptance rate
—
TTFB
1.27 GB
Power draw
250.9 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3-4B-Thinking-2507
Quantization
Q4_K_S
Framework
vLLM
Version
v0.49
Context length
8192 tokens
Batch size
16
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 Radeon AI PRO R9700
Nominal VRAM
32 GB
Measured VRAM (avg)
8.65 GB
OS
macOS 15.0
Driver
amdgpu 6.10
CUDA
N/A
Power draw
250.9 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

    v0.49 · macOS 15.0 · CUDA N/A · 8192 ctx

    158.57 tok/s

    Decode

    583.28 tok/s

    Prefill

    0.64 s

    TTFT

    8.65 GB

    VRAM

    —

    MTP

    250.9 W

    Power draw

    L2 batch seed 2026-10-13: vLLM on hw_id=79 running model_id=57 at Q4_K_S