Arao AI
Back to search results

Qwen3.5-27B · Q3_K_S

Start a discussion
Apple M5 Ultra512GBOllama v0.29L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.5-27B (Q3_K_S) on Apple M5 Ultra with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

84 tok/s

Decode

Decode speed

452.07 tok/s

Prefill

Prefill speed

0.65 s

TTFT

Time to first token

38.96 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

17 days ago

Last verified

Performance

  1. llama.cpp · IQ1_S Decode 131.5 · Prefill 444.06 ·
  2. Ollama · Q3_K_S (current)Decode 84 · Prefill 452.07 ·

Core figures

Decode (avg)
84 tok/s
Prefill (avg)
452.07 tok/s
TTFT (avg)
0.65 s
VRAM (avg)
38.96 GB
MTP acceptance rate
—
TTFB
0.67 GB
Power draw
96.6 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.5-27B
Quantization
Q3_K_S
Framework
Ollama
Version
v0.29
Context length
16384 tokens
Batch size
256
GPU layers
99
Flash Attention
Off

Hardware

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

GPU
Apple M5 Ultra
Nominal VRAM
512 GB
Measured VRAM (avg)
38.96 GB
OS
macOS 26.1
Power draw
96.6 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-19

    v0.29 · macOS 26.1 · 16384 ctx

    84 tok/s

    Decode

    452.07 tok/s

    Prefill

    0.65 s

    TTFT

    38.96 GB

    VRAM

    —

    MTP

    96.6 W

    Power draw

    L2 batch seed 2026-10-19: Ollama on hw_id=29 running model_id=50 at Q3_K_S