Arao AI
Back to search results

gemma-4-26B-A4B-it · IQ3_XS

Start a discussion
Apple M4 Pro64GBvLLM v0.21L2 Cross-framework verified

This page aggregates 1 real-world runs of gemma-4-26B-A4B-it (IQ3_XS) on Apple M4 Pro with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

87.63 tok/s

Decode

Decode speed

428.03 tok/s

Prefill

Prefill speed

0.53 s

TTFT

Time to first token

35.95 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

5 days ago

Last verified

Performance

  1. vLLM · IQ3_XS (current)Decode 87.63 · Prefill 428.03 ·
  2. vLLM · Q4_K_S Decode 72.54 · Prefill 364.75 ·
  3. Ollama · Q5_K_M Decode 56.24 · Prefill 299.52 ·
  4. Ollama · Q5_K_S Decode 59.26 · Prefill 228.14 ·

Core figures

Decode (avg)
87.63 tok/s
Prefill (avg)
428.03 tok/s
TTFT (avg)
0.53 s
VRAM (avg)
35.95 GB
MTP acceptance rate
—
TTFB
0.7 GB
Power draw
69.6 W

Configuration

Member-level fields are taken from the most recent run

Model
gemma-4-26B-A4B-it
Quantization
IQ3_XS
Framework
vLLM
Version
v0.21
Context length
16384 tokens
Batch size
32
GPU layers
999
Flash Attention
On

Hardware

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

GPU
Apple M4 Pro
Nominal VRAM
64 GB
Measured VRAM (avg)
35.95 GB
OS
macOS 15.0
Driver
Apple M4 Pro
CUDA
N/A
Power draw
69.6 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-13

    v0.21 · macOS 15.0 · CUDA N/A · 16384 ctx

    87.63 tok/s

    Decode

    428.03 tok/s

    Prefill

    0.53 s

    TTFT

    35.95 GB

    VRAM

    —

    MTP

    69.6 W

    Power draw

    L2 batch seed 2026-10-13: vLLM on hw_id=25 running model_id=64 at IQ3_XS