Arao AI
Back to search results

Qwen3.6-35B-A3B · Q3_K_L

Start a discussion
Apple M5 Max128GBOllama v0.38L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.6-35B-A3B (Q3_K_L) on Apple M5 Max with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

131.28 tok/s

Decode

Decode speed

555.46 tok/s

Prefill

Prefill speed

0.81 s

TTFT

Time to first token

55.02 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 · IQ3_XS Decode 159.32 · Prefill 706.13 ·
  2. llama.cpp · Q6_K Decode 90.4 · Prefill 328.52 ·
  3. Ollama · Q3_K_L (current)Decode 131.28 · Prefill 555.46 ·
  4. Ollama · Q5_K_S Decode 105.03 · Prefill 472.27 ·

Core figures

Decode (avg)
131.28 tok/s
Prefill (avg)
555.46 tok/s
TTFT (avg)
0.81 s
VRAM (avg)
55.02 GB
MTP acceptance rate
—
TTFB
1.05 GB
Power draw
48.4 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.6-35B-A3B
Quantization
Q3_K_L
Framework
Ollama
Version
v0.38
Context length
4096 tokens
Batch size
256
GPU layers
28
Flash Attention
Off

Hardware

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

GPU
Apple M5 Max
Nominal VRAM
128 GB
Measured VRAM (avg)
55.02 GB
OS
macOS 26.1
Power draw
48.4 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

    v0.38 · macOS 26.1 · 4096 ctx

    131.28 tok/s

    Decode

    555.46 tok/s

    Prefill

    0.81 s

    TTFT

    55.02 GB

    VRAM

    —

    MTP

    48.4 W

    Power draw

    L2 batch seed 2026-10-19: Ollama on hw_id=28 running model_id=47 at Q3_K_L