Arao AI
Back to search results

gemma-4-E2B-it · IQ2_XS

Start a discussion
Apple M2 Ultrallama.cpp v0.25L2 Cross-framework verified

This page aggregates 1 real-world runs of gemma-4-E2B-it (IQ2_XS) on Apple M2 Ultra with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

255.2 tok/s

Decode

Decode speed

1,294.22 tok/s

Prefill

Prefill speed

1.05 s

TTFT

Time to first token

4.12 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

2 days ago

Last verified

Performance

  1. llama.cpp · IQ2_XS (current)Decode 255.2 · Prefill 1294.22 ·
  2. llama.cpp · Q3_K_S Decode 222.24 · Prefill 799.86 ·
  3. Strata · Q3_K_L Decode 291.87 · Prefill 1161.97 ·

Core figures

Decode (avg)
255.2 tok/s
Prefill (avg)
1,294.22 tok/s
TTFT (avg)
1.05 s
VRAM (avg)
4.12 GB
MTP acceptance rate
—
TTFB
2.04 GB
Power draw
68.9 W

Configuration

Member-level fields are taken from the most recent run

Model
gemma-4-E2B-it
Quantization
IQ2_XS
Framework
llama.cpp
Version
v0.25
Context length
32768 tokens
Batch size
512
GPU layers
28
Flash Attention
Off

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)
4.12 GB
OS
macOS 15.0
Driver
Apple M2 Ultra
CUDA
N/A
Power draw
68.9 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-16

    v0.25 · macOS 15.0 · CUDA N/A · 32768 ctx

    255.2 tok/s

    Decode

    1,294.22 tok/s

    Prefill

    1.05 s

    TTFT

    4.12 GB

    VRAM

    —

    MTP

    68.9 W

    Power draw

    L2 batch seed 2026-10-16: llama.cpp on hw_id=103 running model_id=66 at IQ2_XS