Arao AI
Back to search results

Qwen3-4B · Q5_K_M

Start a discussion
Apple M5 Max128GBOllama v0.50L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3-4B (Q5_K_M) on Apple M5 Max with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

194.79 tok/s

Decode

Decode speed

989.8 tok/s

Prefill

Prefill speed

0.12 s

TTFT

Time to first token

9.91 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

5 days ago

Last verified

Performance

  1. llama.cpp · Q3_K_L Decode 253.55 · Prefill 980.61 ·
  2. Ollama · IQ2_XS Decode 308.45 · Prefill 1307.27 ·
  3. Ollama · IQ3_XS Decode 249.13 · Prefill 1132.99 ·
  4. Ollama · Q5_K_M (current)Decode 194.79 · Prefill 989.8 ·
  5. Strata · IQ4_XS Decode 357.04 · Prefill 1403.97 ·

Core figures

Decode (avg)
194.79 tok/s
Prefill (avg)
989.8 tok/s
TTFT (avg)
0.12 s
VRAM (avg)
9.91 GB
MTP acceptance rate
—
TTFB
1.56 GB
Power draw
119.5 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3-4B
Quantization
Q5_K_M
Framework
Ollama
Version
v0.50
Context length
2048 tokens
Batch size
512
GPU layers
33
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)
9.91 GB
OS
macOS 15.0
Driver
Apple M5 Max
CUDA
N/A
Power draw
119.5 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-13

    v0.50 · macOS 15.0 · CUDA N/A · 2048 ctx

    194.79 tok/s

    Decode

    989.8 tok/s

    Prefill

    0.12 s

    TTFT

    9.91 GB

    VRAM

    —

    MTP

    119.5 W

    Power draw

    L2 batch seed 2026-10-13: Ollama on hw_id=28 running model_id=54 at Q5_K_M