Arao AI
Back to search results

Qwen2.5-Omni-3B · Q8_0

Start a discussion
AMD Radeon RX 907016GBvLLM v0.36L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen2.5-Omni-3B (Q8_0) on AMD Radeon RX 9070 with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

91.32 tok/s

Decode

Decode speed

374.63 tok/s

Prefill

Prefill speed

2.39 s

TTFT

Time to first token

9.35 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

5 days ago

Last verified

Performance

  1. vLLM · Q4_K_S Decode 147.81 · Prefill 510.18 ·
  2. vLLM · Q8_0 (current)Decode 91.32 · Prefill 374.63 ·
  3. Strata · Q6_K Decode 126.97 · Prefill 623.22 ·

Core figures

Decode (avg)
91.32 tok/s
Prefill (avg)
374.63 tok/s
TTFT (avg)
2.39 s
VRAM (avg)
9.35 GB
MTP acceptance rate
—
TTFB
0.73 GB
Power draw
309.7 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen2.5-Omni-3B
Quantization
Q8_0
Framework
vLLM
Version
v0.36
Context length
2048 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
AMD Radeon RX 9070
Nominal VRAM
16 GB
Measured VRAM (avg)
9.35 GB
OS
macOS 15.0
Driver
amdgpu-pro 24.20
CUDA
N/A
Power draw
309.7 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-13

    v0.36 · macOS 15.0 · CUDA N/A · 2048 ctx

    91.32 tok/s

    Decode

    374.63 tok/s

    Prefill

    2.39 s

    TTFT

    9.35 GB

    VRAM

    —

    MTP

    309.7 W

    Power draw

    L2 batch seed 2026-10-13: vLLM on hw_id=21 running model_id=60 at Q8_0