Arao AI
Back to search results

Qwen3.6-27B · Q3_K_M

Start a discussion
Apple M4 Max128GBvLLM v0.33L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.6-27B (Q3_K_M) on Apple M4 Max with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

174.24 tok/s

Decode

Decode speed

673.46 tok/s

Prefill

Prefill speed

0.17 s

TTFT

Time to first token

40.66 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. llama.cpp · IQ2_M Decode 189.88 · Prefill 951.61 ·
  2. llama.cpp · IQ2_XS Decode 191.16 · Prefill 1032.54 ·
  3. vLLM · Q3_K_M (current)Decode 174.24 · Prefill 673.46 ·
  4. Ollama · Q6_K Decode 99.52 · Prefill 397.09 ·

Core figures

Decode (avg)
174.24 tok/s
Prefill (avg)
673.46 tok/s
TTFT (avg)
0.17 s
VRAM (avg)
40.66 GB
MTP acceptance rate
—
TTFB
1.39 GB
Power draw
94.7 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.6-27B
Quantization
Q3_K_M
Framework
vLLM
Version
v0.33
Context length
16384 tokens
Batch size
8
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
Apple M4 Max
Nominal VRAM
128 GB
Measured VRAM (avg)
40.66 GB
OS
macOS 15.0
Driver
Apple M4 Max
CUDA
N/A
Power draw
94.7 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 2026-10-10

    v0.33 · macOS 15.0 · CUDA N/A · 16384 ctx

    174.24 tok/s

    Decode

    673.46 tok/s

    Prefill

    0.17 s

    TTFT

    40.66 GB

    VRAM

    —

    MTP

    94.7 W

    Power draw

    L2 batch seed 2026-10-10: vLLM on hw_id=26 running model_id=6 at Q3_K_M