Arao AI
Back to search results

gemma-4-E4B-it · Q3_K_L

Start a discussion
Apple M5 Max128GBStrata v0.22L2 Cross-framework verified

This page aggregates 1 real-world runs of gemma-4-E4B-it (Q3_K_L) on Apple M5 Max with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.

374.77 tok/s

Decode

Decode speed

1,948.71 tok/s

Prefill

Prefill speed

0.83 s

TTFT

Time to first token

8.15 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 Decode 155.09 · Prefill 679.29 ·
  4. Strata · Q3_K_L (current)Decode 374.77 · Prefill 1948.71 ·

Core figures

Decode (avg)
374.77 tok/s
Prefill (avg)
1,948.71 tok/s
TTFT (avg)
0.83 s
VRAM (avg)
8.15 GB
MTP acceptance rate
0.68
TTFB
3 GB
Power draw
158 W

Configuration

Member-level fields are taken from the most recent run

Model
gemma-4-E4B-it
Quantization
Q3_K_L
Framework
Strata
Version
v0.22
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 M5 Max
Nominal VRAM
128 GB
Measured VRAM (avg)
8.15 GB
OS
Ubuntu 22.04
Driver
560.94
CUDA
12.4
Power draw
158 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

    v0.22 · Ubuntu 22.04 · CUDA 12.4 · 8192 ctx

    374.77 tok/s

    Decode

    1,948.71 tok/s

    Prefill

    0.83 s

    TTFT

    8.15 GB

    VRAM

    0.68

    MTP

    158 W

    Power draw

    L2 batch seed 2026-10-16: Strata on hw_id=28 running model_id=67 at Q3_K_L