Arao AI
Back to search results

Qwen3-8B · Q2_K

Start a discussion
Apple M2 UltraOllama v0.53L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3-8B (Q2_K) on Apple M2 Ultra with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

112.25 tok/s

Decode

Decode speed

576.42 tok/s

Prefill

Prefill speed

1.28 s

TTFT

Time to first token

11.34 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

16 days ago

Last verified

Performance

  1. llama.cpp · Q3_K_S Decode 119.1 · Prefill 434.44 ·
  2. Ollama · Q2_K (current)Decode 112.25 · Prefill 576.42 ·
  3. llama.cpp · Q4_K_S Decode 108.67 · Prefill 436.07 ·
  4. llama.cpp · Q5_K_S Decode 87.67 · Prefill 476.02 ·
  5. Ollama · Q5_K_M Decode 79.98 · Prefill 403.29 ·
  6. Ollama · Q8_0 Decode 67.83 · Prefill 284.74 ·

Core figures

Decode (avg)
112.25 tok/s
Prefill (avg)
576.42 tok/s
TTFT (avg)
1.28 s
VRAM (avg)
11.34 GB
MTP acceptance rate
—
TTFB
0.9 GB
Power draw
101.3 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3-8B
Quantization
Q2_K
Framework
Ollama
Version
v0.53
Context length
4096 tokens
Batch size
256
GPU layers
99
Flash Attention
On

Hardware

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

GPU
Apple M2 Ultra
Measured VRAM (avg)
11.34 GB
OS
macOS 26.1
Power draw
101.3 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

    v0.53 · macOS 26.1 · 4096 ctx

    112.25 tok/s

    Decode

    576.42 tok/s

    Prefill

    1.28 s

    TTFT

    11.34 GB

    VRAM

    —

    MTP

    101.3 W

    Power draw

    L2 batch seed 2026-10-19: Ollama on hw_id=103 running model_id=53 at Q2_K