Arao AI
Back to search results

gemma-4-E4B-it · FP16

Start a discussion
Apple M5 Max128GBStrata v0.53L2 Cross-framework verified

This page aggregates 1 real-world runs of gemma-4-E4B-it (FP16) on Apple M5 Max with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.

155.09 tok/s

Decode

Decode speed

679.29 tok/s

Prefill

Prefill speed

0.3 s

TTFT

Time to first token

17.64 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

2 days ago

Last verified

Performance

  1. Ollama · IQ3_XS Decode 264.68 · Prefill 1192.68 ·
  2. Ollama · Q2_K Decode 303.01 · Prefill 1437.6 ·
  3. Strata · FP16 (current)Decode 155.09 · Prefill 679.29 ·
  4. Strata · Q3_K_L Decode 374.77 · Prefill 1948.71 ·

Core figures

Decode (avg)
155.09 tok/s
Prefill (avg)
679.29 tok/s
TTFT (avg)
0.3 s
VRAM (avg)
17.64 GB
MTP acceptance rate
0.47
TTFB
1.24 GB
Power draw
276.7 W

Configuration

Member-level fields are taken from the most recent run

Model
gemma-4-E4B-it
Quantization
FP16
Framework
Strata
Version
v0.53
Context length
16384 tokens
Batch size
256
GPU layers
12
Flash Attention
Off

Hardware

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

GPU
Apple M5 Max
Nominal VRAM
128 GB
Measured VRAM (avg)
17.64 GB
OS
Ubuntu 22.04
Driver
565.57
CUDA
12.4
Power draw
276.7 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

    v0.53 · Ubuntu 22.04 · CUDA 12.4 · 16384 ctx

    155.09 tok/s

    Decode

    679.29 tok/s

    Prefill

    0.3 s

    TTFT

    17.64 GB

    VRAM

    0.47

    MTP

    276.7 W

    Power draw

    L2 batch seed 2026-10-16: Strata on hw_id=28 running model_id=67 at FP16