Arao AI
Back to search results

Qwen2.5-Omni-3B · IQ2_XS

Start a discussion
Qualcomm Snapdragon X ElitevLLM v0.46L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen2.5-Omni-3B (IQ2_XS) on Qualcomm Snapdragon X Elite with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

103.58 tok/s

Decode

Decode speed

468.29 tok/s

Prefill

Prefill speed

1.41 s

TTFT

Time to first token

5.12 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

2 days ago

Last verified

Performance

  1. vLLM · IQ2_XS (current)Decode 103.58 · Prefill 468.29 ·
  2. vLLM · Q5_K_S Decode 65.67 · Prefill 300.03 ·
  3. Ollama · Q2_K Decode 89.58 · Prefill 387.16 ·
  4. Ollama · Q3_K_S Decode 76.04 · Prefill 366.79 ·

Core figures

Decode (avg)
103.58 tok/s
Prefill (avg)
468.29 tok/s
TTFT (avg)
1.41 s
VRAM (avg)
5.12 GB
MTP acceptance rate
—
TTFB
0.83 GB
Power draw
9.3 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen2.5-Omni-3B
Quantization
IQ2_XS
Framework
vLLM
Version
v0.46
Context length
4096 tokens
Batch size
8
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
Qualcomm Snapdragon X Elite
Measured VRAM (avg)
5.12 GB
OS
Windows 11 23H2
Driver
Qualcomm Adreno 7.0
CUDA
N/A
Power draw
9.3 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-16

    v0.46 · Windows 11 23H2 · CUDA N/A · 4096 ctx

    103.58 tok/s

    Decode

    468.29 tok/s

    Prefill

    1.41 s

    TTFT

    5.12 GB

    VRAM

    —

    MTP

    9.3 W

    Power draw

    L2 batch seed 2026-10-16: vLLM on hw_id=104 running model_id=60 at IQ2_XS