Arao AI
Back to search results

Qwen2.5-Omni-3B · IQ4_XS

Start a discussion
Intel Arc A770 16GB16GBvLLM v0.26L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen2.5-Omni-3B (IQ4_XS) on Intel Arc A770 16GB with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

86.78 tok/s

Decode

Decode speed

278.58 tok/s

Prefill

Prefill speed

2.2 s

TTFT

Time to first token

6.76 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

2 days ago

Last verified

Performance

  1. llama.cpp · IQ2_XS Decode 113.09 · Prefill 493.14 ·
  2. vLLM · IQ2_M Decode 105.24 · Prefill 529.61 ·
  3. vLLM · IQ4_XS (current)Decode 86.78 · Prefill 278.58 ·
  4. llama.cpp · Q3_K_S Decode 92.22 · Prefill 373.8 ·
  5. Strata · IQ2_M Decode 145.2 · Prefill 469.28 ·

Core figures

Decode (avg)
86.78 tok/s
Prefill (avg)
278.58 tok/s
TTFT (avg)
2.2 s
VRAM (avg)
6.76 GB
MTP acceptance rate
—
TTFB
0.69 GB
Power draw
154.7 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen2.5-Omni-3B
Quantization
IQ4_XS
Framework
vLLM
Version
v0.26
Context length
4096 tokens
Batch size
32
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
Intel Arc A770 16GB
Nominal VRAM
16 GB
Measured VRAM (avg)
6.76 GB
OS
Ubuntu 24.04
Driver
Intel Arc 31.0.101.5186
CUDA
12.6
Power draw
154.7 W

Sources & evidence

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

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

    v0.26 · Ubuntu 24.04 · CUDA 12.6 · 4096 ctx

    86.78 tok/s

    Decode

    278.58 tok/s

    Prefill

    2.2 s

    TTFT

    6.76 GB

    VRAM

    —

    MTP

    154.7 W

    Power draw

    L2 batch seed 2026-10-16: vLLM on hw_id=97 running model_id=60 at IQ4_XS