Arao AI
Back to search results

Qwen3.8-9B · IQ1_S

Start a discussion
NVIDIA RTX 4080 Super16GBllama.cpp v0.39L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.8-9B (IQ1_S) on NVIDIA RTX 4080 Super with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

204.16 tok/s

Decode

Decode speed

1,111.21 tok/s

Prefill

Prefill speed

0.65 s

TTFT

Time to first token

10.04 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. llama.cpp · IQ1_S (current)Decode 204.16 · Prefill 1111.21 ·
  2. Ollama · IQ2_XS Decode 174.28 · Prefill 597.11 ·
  3. Ollama · Q3_K_M Decode 141.22 · Prefill 455.03 ·

Core figures

Decode (avg)
204.16 tok/s
Prefill (avg)
1,111.21 tok/s
TTFT (avg)
0.65 s
VRAM (avg)
10.04 GB
MTP acceptance rate
—
TTFB
1.63 GB
Power draw
253.7 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.8-9B
Quantization
IQ1_S
Framework
llama.cpp
Version
v0.39
Context length
4096 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
NVIDIA RTX 4080 Super
Nominal VRAM
16 GB
Measured VRAM (avg)
10.04 GB
OS
Ubuntu 24.04
Driver
572.83
CUDA
12.8
Power draw
253.7 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

    v0.39 · Ubuntu 24.04 · CUDA 12.8 · 4096 ctx

    204.16 tok/s

    Decode

    1,111.21 tok/s

    Prefill

    0.65 s

    TTFT

    10.04 GB

    VRAM

    —

    MTP

    253.7 W

    Power draw

    L2 batch seed 2026-10-18: llama.cpp on hw_id=20 running model_id=52 at IQ1_S