Arao AI
Back to search results

Qwen3.5-0.8B · FP16

Start a discussion
Apple M2 UltraOllama v0.32L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.5-0.8B (FP16) on Apple M2 Ultra with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

117 tok/s

Decode

Decode speed

552.38 tok/s

Prefill

Prefill speed

0.68 s

TTFT

Time to first token

5.21 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 344.97 · Prefill 1343.81 ·
  2. llama.cpp · Q6_K Decode 188.49 · Prefill 853.9 ·
  3. Ollama · FP16 (current)Decode 117 · Prefill 552.38 ·
  4. Ollama · Q8_0 Decode 150.72 · Prefill 712.42 ·
  5. Ollama · IQ4_XS Decode 235.75 · Prefill 979.9 ·
  6. Strata · Q3_K_L Decode 328.49 · Prefill 1202.75 ·
  7. Strata · Q8_0 Decode 256.65 · Prefill 954.91 ·

Core figures

Decode (avg)
117 tok/s
Prefill (avg)
552.38 tok/s
TTFT (avg)
0.68 s
VRAM (avg)
5.21 GB
MTP acceptance rate
—
TTFB
0.94 GB
Power draw
117.9 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.5-0.8B
Quantization
FP16
Framework
Ollama
Version
v0.32
Context length
8192 tokens
Batch size
256
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 M2 Ultra
Measured VRAM (avg)
5.21 GB
OS
macOS 26.1
Power draw
117.9 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-19

    v0.32 · macOS 26.1 · 8192 ctx

    117 tok/s

    Decode

    552.38 tok/s

    Prefill

    0.68 s

    TTFT

    5.21 GB

    VRAM

    —

    MTP

    117.9 W

    Power draw

    L2 batch seed 2026-10-19: Ollama on hw_id=103 running model_id=51 at FP16