Arao AI
Back to search results

Qwen3.8-9B · Q3_K_S

Start a discussion
NVIDIA RTX 309024GBllama.cpp v0.21L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.8-9B (Q3_K_S) on NVIDIA RTX 3090 with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

130.89 tok/s

Decode

Decode speed

575.17 tok/s

Prefill

Prefill speed

0.35 s

TTFT

Time to first token

14.38 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. llama.cpp · Q3_K_S (current)Decode 130.89 · Prefill 575.17 ·
  2. Ollama · IQ4_XS Decode 98.13 · Prefill 496.33 ·

Core figures

Decode (avg)
130.89 tok/s
Prefill (avg)
575.17 tok/s
TTFT (avg)
0.35 s
VRAM (avg)
14.38 GB
MTP acceptance rate
—
TTFB
1.05 GB
Power draw
343.6 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.8-9B
Quantization
Q3_K_S
Framework
llama.cpp
Version
v0.21
Context length
32768 tokens
Batch size
256
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
NVIDIA RTX 3090
Nominal VRAM
24 GB
Measured VRAM (avg)
14.38 GB
OS
Windows 11 23H2
Driver
560.94
CUDA
12.4
Power draw
343.6 W

Sources & evidence

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

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

    v0.21 · Windows 11 23H2 · CUDA 12.4 · 32768 ctx

    130.89 tok/s

    Decode

    575.17 tok/s

    Prefill

    0.35 s

    TTFT

    14.38 GB

    VRAM

    —

    MTP

    343.6 W

    Power draw

    L2 batch seed 2026-10-10: llama.cpp on hw_id=2 running model_id=52 at Q3_K_S