Arao AI
Back to search results

Qwen3.8-9B · Q3_K_M

Start a discussion
AMD Radeon RX 7900 XTX24GBvLLM v0.48L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.8-9B (Q3_K_M) on AMD Radeon RX 7900 XTX with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

85.52 tok/s

Decode

Decode speed

281.83 tok/s

Prefill

Prefill speed

1.43 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

Today

Last verified

Performance

  1. vLLM · Q3_K_L Decode 79.57 · Prefill 292.41 ·
  2. vLLM · Q3_K_M (current)Decode 85.52 · Prefill 281.83 ·
  3. Ollama · Q4_K_M Decode 61.26 · Prefill 253.43 ·
  4. Ollama · Q5_K_M Decode 56.51 · Prefill 294.11 ·
  5. llama.cpp · IQ2_M Decode 116.92 · Prefill 403.5 ·
  6. llama.cpp · Q4_K_S Decode 92.32 · Prefill 411.04 ·
  7. vLLM · Q8_0 Decode 60.74 · Prefill 270.33 ·

Core figures

Decode (avg)
85.52 tok/s
Prefill (avg)
281.83 tok/s
TTFT (avg)
1.43 s
VRAM (avg)
14.95 GB
MTP acceptance rate
—
TTFB
0.68 GB
Power draw
233.9 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.8-9B
Quantization
Q3_K_M
Framework
vLLM
Version
v0.48
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 RX 7900 XTX
Nominal VRAM
24 GB
Measured VRAM (avg)
14.95 GB
OS
Ubuntu 22.04
Driver
amdgpu-pro 24.20
CUDA
12.4
Power draw
233.9 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-18

    v0.48 · Ubuntu 22.04 · CUDA 12.4 · 8192 ctx

    85.52 tok/s

    Decode

    281.83 tok/s

    Prefill

    1.43 s

    TTFT

    14.95 GB

    VRAM

    —

    MTP

    233.9 W

    Power draw

    L2 batch seed 2026-10-18: vLLM on hw_id=11 running model_id=52 at Q3_K_M