Arao AI
Back to search results

Qwen3-8B · IQ2_M

Start a discussion
Apple M5 Max128GBOllama v0.33L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3-8B (IQ2_M) on Apple M5 Max with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

82.41 tok/s

Decode

Decode speed

283.9 tok/s

Prefill

Prefill speed

2.19 s

TTFT

Time to first token

10.84 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 · IQ1_S Decode 95.3 · Prefill 467.44 ·
  2. Ollama · IQ2_M (current)Decode 82.41 · Prefill 283.9 ·
  3. Ollama · Q6_K Decode 50.37 · Prefill 248.31 ·
  4. llama.cpp · IQ3_XS Decode 233.71 · Prefill 783.62 ·
  5. Ollama · IQ3_XS Decode 212.17 · Prefill 767.04 ·
  6. Ollama · Q4_K_S Decode 185.61 · Prefill 631.22 ·
  7. llama.cpp · Q6_K Decode 137.3 · Prefill 696.86 ·
  8. Strata · Q4_K_M Decode 252.52 · Prefill 1255.16 ·

Core figures

Decode (avg)
82.41 tok/s
Prefill (avg)
283.9 tok/s
TTFT (avg)
2.19 s
VRAM (avg)
10.84 GB
MTP acceptance rate
—
TTFB
0.66 GB
Power draw
42.3 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3-8B
Quantization
IQ2_M
Framework
Ollama
Version
v0.33
Context length
8192 tokens
Batch size
256
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)
10.84 GB
OS
macOS 26.1
Power draw
42.3 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

    v0.33 · macOS 26.1 · 8192 ctx

    82.41 tok/s

    Decode

    283.9 tok/s

    Prefill

    2.19 s

    TTFT

    10.84 GB

    VRAM

    —

    MTP

    42.3 W

    Power draw

    L2 batch seed 2026-10-19: Ollama on hw_id=28 running model_id=53 at IQ2_M