Arao AI
Back to search results

Qwen3.6-35B-A3B · IQ2_XS

Start a discussion
Apple M3 Max36GBllama.cpp v0.25L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.6-35B-A3B (IQ2_XS) on Apple M3 Max with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

89.41 tok/s

Decode

Decode speed

406.89 tok/s

Prefill

Prefill speed

0.17 s

TTFT

Time to first token

34.2 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_XS (current)Decode 89.41 · Prefill 406.89 ·
  2. vLLM · IQ1_S Decode 102.56 · Prefill 411.25 ·

Core figures

Decode (avg)
89.41 tok/s
Prefill (avg)
406.89 tok/s
TTFT (avg)
0.17 s
VRAM (avg)
34.2 GB
MTP acceptance rate
—
TTFB
0.72 GB
Power draw
90.9 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.6-35B-A3B
Quantization
IQ2_XS
Framework
llama.cpp
Version
v0.25
Context length
16384 tokens
Batch size
512
GPU layers
28
Flash Attention
On

Hardware

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

GPU
Apple M3 Max
Nominal VRAM
36 GB
Measured VRAM (avg)
34.2 GB
OS
macOS 15.0
Driver
Apple M3 Max
CUDA
N/A
Power draw
90.9 W

Sources & evidence

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

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

    v0.25 · macOS 15.0 · CUDA N/A · 16384 ctx

    89.41 tok/s

    Decode

    406.89 tok/s

    Prefill

    0.17 s

    TTFT

    34.2 GB

    VRAM

    —

    MTP

    90.9 W

    Power draw

    L2 batch seed 2026-10-10: llama.cpp on hw_id=4 running model_id=47 at IQ2_XS