Arao AI
Back to search results

Qwen3.8-9B · Q2_K

Start a discussion
Apple M3 Max36GBStrata v0.49L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.8-9B (Q2_K) on Apple M3 Max with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.

180.32 tok/s

Decode

Decode speed

900.55 tok/s

Prefill

Prefill speed

1.84 s

TTFT

Time to first token

12.5 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. llama.cpp · IQ2_M Decode 136.47 · Prefill 438.69 ·
  2. llama.cpp · IQ3_S Decode 126.59 · Prefill 664.89 ·
  3. Strata · Q2_K (current)Decode 180.32 · Prefill 900.55 ·

Core figures

Decode (avg)
180.32 tok/s
Prefill (avg)
900.55 tok/s
TTFT (avg)
1.84 s
VRAM (avg)
12.5 GB
MTP acceptance rate
0.52
TTFB
1.44 GB
Power draw
40.5 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.8-9B
Quantization
Q2_K
Framework
Strata
Version
v0.49
Context length
8192 tokens
Batch size
512
GPU layers
15
Flash Attention
On

Hardware

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

GPU
Apple M3 Max
Nominal VRAM
36 GB
Measured VRAM (avg)
12.5 GB
OS
macOS 15.0
Driver
Apple M3 Max
CUDA
N/A
Power draw
40.5 W

Sources & evidence

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

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

    v0.49 · macOS 15.0 · CUDA N/A · 8192 ctx

    180.32 tok/s

    Decode

    900.55 tok/s

    Prefill

    1.84 s

    TTFT

    12.5 GB

    VRAM

    0.52

    MTP

    40.5 W

    Power draw

    L2 batch seed 2026-10-10: Strata on hw_id=4 running model_id=52 at Q2_K