Arao AI
Back to search results

Qwen3.8-9B · Q3_K_M

Start a discussion
NVIDIA RTX 508016GBllama.cpp v0.29L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.8-9B (Q3_K_M) on NVIDIA RTX 5080 with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

108.92 tok/s

Decode

Decode speed

455.22 tok/s

Prefill

Prefill speed

1.01 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. llama.cpp · IQ4_XS Decode 97.14 · Prefill 496.38 ·
  2. llama.cpp · Q3_K_M (current)Decode 108.92 · Prefill 455.22 ·
  3. vLLM · IQ2_M Decode 131.31 · Prefill 559.09 ·

Core figures

Decode (avg)
108.92 tok/s
Prefill (avg)
455.22 tok/s
TTFT (avg)
1.01 s
VRAM (avg)
14.95 GB
MTP acceptance rate
—
TTFB
0.87 GB
Power draw
262.5 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.8-9B
Quantization
Q3_K_M
Framework
llama.cpp
Version
v0.29
Context length
4096 tokens
Batch size
256
GPU layers
99
Flash Attention
Off

Hardware

Nominal and measured figures are shown side by side; whether it runs is the reader's call

GPU
NVIDIA RTX 5080
Nominal VRAM
16 GB
Measured VRAM (avg)
14.95 GB
OS
Ubuntu 24.04
Driver
560.94
CUDA
12.8
Power draw
262.5 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

    v0.29 · Ubuntu 24.04 · CUDA 12.8 · 4096 ctx

    108.92 tok/s

    Decode

    455.22 tok/s

    Prefill

    1.01 s

    TTFT

    14.95 GB

    VRAM

    —

    MTP

    262.5 W

    Power draw

    L2 batch seed 2026-10-10: llama.cpp on hw_id=14 running model_id=52 at Q3_K_M