Arao AI
Back to search results

gemma-4-E4B-it · Q3_K_S

Start a discussion
NVIDIA RTX 509032GBvLLM v0.45L2 Cross-framework verified

This page aggregates 1 real-world runs of gemma-4-E4B-it (Q3_K_S) on NVIDIA RTX 5090 with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

310.95 tok/s

Decode

Decode speed

1,691.37 tok/s

Prefill

Prefill speed

1.24 s

TTFT

Time to first token

7.56 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

2 days ago

Last verified

Performance

  1. vLLM · Q3_K_L Decode 336.48 · Prefill 1309.15 ·
  2. vLLM · Q3_K_S (current)Decode 310.95 · Prefill 1691.37 ·
  3. Strata · Q3_K_S Decode 326.07 · Prefill 1598 ·
  4. Strata · Q5_K_M Decode 361.26 · Prefill 1894.18 ·

Core figures

Decode (avg)
310.95 tok/s
Prefill (avg)
1,691.37 tok/s
TTFT (avg)
1.24 s
VRAM (avg)
7.56 GB
MTP acceptance rate
—
TTFB
2.49 GB
Power draw
419.5 W

Configuration

Member-level fields are taken from the most recent run

Model
gemma-4-E4B-it
Quantization
Q3_K_S
Framework
vLLM
Version
v0.45
Context length
2048 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
NVIDIA RTX 5090
Nominal VRAM
32 GB
Measured VRAM (avg)
7.56 GB
OS
Ubuntu 22.04
Driver
570.86
CUDA
12.6
Power draw
419.5 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

    v0.45 · Ubuntu 22.04 · CUDA 12.6 · 2048 ctx

    310.95 tok/s

    Decode

    1,691.37 tok/s

    Prefill

    1.24 s

    TTFT

    7.56 GB

    VRAM

    —

    MTP

    419.5 W

    Power draw

    L2 batch seed 2026-10-16: vLLM on hw_id=13 running model_id=67 at Q3_K_S