Arao AI
Back to search results

Qwen2.5-0.5B-Instruct · Q8_0

Start a discussion
Apple M5 Pro64GBvLLM v0.39L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen2.5-0.5B-Instruct (Q8_0) on Apple M5 Pro with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

284.14 tok/s

Decode

Decode speed

1,555.29 tok/s

Prefill

Prefill speed

0.9 s

TTFT

Time to first token

3.31 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 · Q5_K_S Decode 376.55 · Prefill 1237.86 ·
  2. vLLM · Q8_0 (current)Decode 284.14 · Prefill 1555.29 ·

Core figures

Decode (avg)
284.14 tok/s
Prefill (avg)
1,555.29 tok/s
TTFT (avg)
0.9 s
VRAM (avg)
3.31 GB
MTP acceptance rate
—
TTFB
2.27 GB
Power draw
117.3 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen2.5-0.5B-Instruct
Quantization
Q8_0
Framework
vLLM
Version
v0.39
Context length
16384 tokens
Batch size
32
GPU layers
999
Flash Attention
Off

Hardware

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

GPU
Apple M5 Pro
Nominal VRAM
64 GB
Measured VRAM (avg)
3.31 GB
OS
macOS 15.0
Driver
Apple M5 Pro
CUDA
N/A
Power draw
117.3 W

Sources & evidence

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

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

    v0.39 · macOS 15.0 · CUDA N/A · 16384 ctx

    284.14 tok/s

    Decode

    1,555.29 tok/s

    Prefill

    0.9 s

    TTFT

    3.31 GB

    VRAM

    —

    MTP

    117.3 W

    Power draw

    L2 batch seed 2026-10-13: vLLM on hw_id=27 running model_id=59 at Q8_0