Arao AI
Back to search results

Qwen3.8-9B · Q4_K_S

Start a discussion
NVIDIA RTX PRO 6000 Blackwell 96GB96GBvLLM v0.39L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.8-9B (Q4_K_S) on NVIDIA RTX PRO 6000 Blackwell 96GB with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

212.78 tok/s

Decode

Decode speed

1,065.77 tok/s

Prefill

Prefill speed

0.48 s

TTFT

Time to first token

16.84 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

16 days ago

Last verified

Performance

  1. vLLM · Q4_K_S (current)Decode 212.78 · Prefill 1065.77 ·
  2. Ollama · Q3_K_M Decode 221.75 · Prefill 855.6 ·
  3. Ollama · Q5_K_M Decode 148.52 · Prefill 762.34 ·

Core figures

Decode (avg)
212.78 tok/s
Prefill (avg)
1,065.77 tok/s
TTFT (avg)
0.48 s
VRAM (avg)
16.84 GB
MTP acceptance rate
—
TTFB
1.7 GB
Power draw
317.6 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.8-9B
Quantization
Q4_K_S
Framework
vLLM
Version
v0.39
Context length
2048 tokens
Batch size
32
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
NVIDIA RTX PRO 6000 Blackwell 96GB
Nominal VRAM
96 GB
Measured VRAM (avg)
16.84 GB
OS
Ubuntu 24.04
Driver
572.83
CUDA
12.6
Power draw
317.6 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

    v0.39 · Ubuntu 24.04 · CUDA 12.6 · 2048 ctx

    212.78 tok/s

    Decode

    1,065.77 tok/s

    Prefill

    0.48 s

    TTFT

    16.84 GB

    VRAM

    —

    MTP

    317.6 W

    Power draw

    L2 batch seed 2026-10-19: vLLM on hw_id=68 running model_id=52 at Q4_K_S