Arao AI
Back to search results

Qwen3.8-9B · IQ2_M

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

This page aggregates 1 real-world runs of Qwen3.8-9B (IQ2_M) on NVIDIA RTX 5080 with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

131.31 tok/s

Decode

Decode speed

559.09 tok/s

Prefill

Prefill speed

1.55 s

TTFT

Time to first token

11.93 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 Decode 108.92 · Prefill 455.22 ·
  3. vLLM · IQ2_M (current)Decode 131.31 · Prefill 559.09 ·

Core figures

Decode (avg)
131.31 tok/s
Prefill (avg)
559.09 tok/s
TTFT (avg)
1.55 s
VRAM (avg)
11.93 GB
MTP acceptance rate
—
TTFB
1.05 GB
Power draw
312.7 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.8-9B
Quantization
IQ2_M
Framework
vLLM
Version
v0.29
Context length
2048 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
NVIDIA RTX 5080
Nominal VRAM
16 GB
Measured VRAM (avg)
11.93 GB
OS
Windows 11 23H2
Driver
570.86
CUDA
12.8
Power draw
312.7 W

Sources & evidence

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

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

    v0.29 · Windows 11 23H2 · CUDA 12.8 · 2048 ctx

    131.31 tok/s

    Decode

    559.09 tok/s

    Prefill

    1.55 s

    TTFT

    11.93 GB

    VRAM

    —

    MTP

    312.7 W

    Power draw

    L2 batch seed 2026-10-10: vLLM on hw_id=14 running model_id=52 at IQ2_M