Arao AI
Back to search results

Qwen3.5-35B-A3B · IQ2_XS

Start a discussion
Apple M4 Max128GBOllama v0.39L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.5-35B-A3B (IQ2_XS) on Apple M4 Max with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

138.01 tok/s

Decode

Decode speed

631.53 tok/s

Prefill

Prefill speed

2.31 s

TTFT

Time to first token

38.39 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. vLLM · IQ4_XS Decode 115.28 · Prefill 541.34 ·
  2. vLLM · Q4_K_S Decode 115.9 · Prefill 629 ·
  3. Ollama · IQ2_XS (current)Decode 138.01 · Prefill 631.53 ·

Core figures

Decode (avg)
138.01 tok/s
Prefill (avg)
631.53 tok/s
TTFT (avg)
2.31 s
VRAM (avg)
38.39 GB
MTP acceptance rate
—
TTFB
1.1 GB
Power draw
39.3 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.5-35B-A3B
Quantization
IQ2_XS
Framework
Ollama
Version
v0.39
Context length
8192 tokens
Batch size
256
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 M4 Max
Nominal VRAM
128 GB
Measured VRAM (avg)
38.39 GB
OS
macOS 15.0
Driver
Apple M4 Max
CUDA
N/A
Power draw
39.3 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

    v0.39 · macOS 15.0 · CUDA N/A · 8192 ctx

    138.01 tok/s

    Decode

    631.53 tok/s

    Prefill

    2.31 s

    TTFT

    38.39 GB

    VRAM

    —

    MTP

    39.3 W

    Power draw

    L2 batch seed 2026-10-10: Ollama on hw_id=26 running model_id=15 at IQ2_XS