Arao AI
Back to search results

gemma-4-31B-it · Q2_K

Start a discussion
Apple M3 Max36GBllama.cpp v0.39L2 Cross-framework verified

This page aggregates 1 real-world runs of gemma-4-31B-it (Q2_K) on Apple M3 Max with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

64.47 tok/s

Decode

Decode speed

330.47 tok/s

Prefill

Prefill speed

1.47 s

TTFT

Time to first token

34.2 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. llama.cpp · Q2_K (current)Decode 64.47 · Prefill 330.47 ·
  2. Ollama · Q2_K Decode 65.74 · Prefill 261.65 ·

Core figures

Decode (avg)
64.47 tok/s
Prefill (avg)
330.47 tok/s
TTFT (avg)
1.47 s
VRAM (avg)
34.2 GB
MTP acceptance rate
—
TTFB
0.52 GB
Power draw
42 W

Configuration

Member-level fields are taken from the most recent run

Model
gemma-4-31B-it
Quantization
Q2_K
Framework
llama.cpp
Version
v0.39
Context length
16384 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 M3 Max
Nominal VRAM
36 GB
Measured VRAM (avg)
34.2 GB
OS
macOS 15.0
Driver
Apple M3 Max
CUDA
N/A
Power draw
42 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

    v0.39 · macOS 15.0 · CUDA N/A · 16384 ctx

    64.47 tok/s

    Decode

    330.47 tok/s

    Prefill

    1.47 s

    TTFT

    34.2 GB

    VRAM

    —

    MTP

    42 W

    Power draw

    L2 batch seed 2026-10-10: llama.cpp on hw_id=4 running model_id=63 at Q2_K