Arao AI
Back to search results

Qwen3.6-27B · IQ3_S

Start a discussion
Apple M5 Ultra512GBOllama v0.28L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.6-27B (IQ3_S) on Apple M5 Ultra with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

85.88 tok/s

Decode

Decode speed

388.53 tok/s

Prefill

Prefill speed

0.99 s

TTFT

Time to first token

39.52 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

16 days ago

Last verified

Performance

  1. llama.cpp · IQ3_XS Decode 91.11 · Prefill 454.84 ·
  2. llama.cpp · Q4_K_S Decode 81.06 · Prefill 387.21 ·
  3. Ollama · IQ3_S (current)Decode 85.88 · Prefill 388.53 ·
  4. Ollama · Q6_K Decode 53.29 · Prefill 218.93 ·

Core figures

Decode (avg)
85.88 tok/s
Prefill (avg)
388.53 tok/s
TTFT (avg)
0.99 s
VRAM (avg)
39.52 GB
MTP acceptance rate
—
TTFB
0.69 GB
Power draw
117.4 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.6-27B
Quantization
IQ3_S
Framework
Ollama
Version
v0.28
Context length
16384 tokens
Batch size
512
GPU layers
20
Flash Attention
On

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)
39.52 GB
OS
macOS 26.1
Power draw
117.4 W

Get started

Original model resources

Base model resources may not match this quantization. Use the verified deployment weights above when available.

Sources & evidence

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

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

    v0.28 · macOS 26.1 · 16384 ctx

    85.88 tok/s

    Decode

    388.53 tok/s

    Prefill

    0.99 s

    TTFT

    39.52 GB

    VRAM

    —

    MTP

    117.4 W

    Power draw

    L2 batch seed 2026-10-19: Ollama on hw_id=29 running model_id=6 at IQ3_S