Arao AI
Back to search results

Ministral-3-8B-Instruct-2512 · Q4_K_S

Start a discussion
NVIDIA RTX 309024GBvLLM v0.34L2 Cross-framework verified

This page aggregates 1 real-world runs of Ministral-3-8B-Instruct-2512 (Q4_K_S) on NVIDIA RTX 3090 with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

94.16 tok/s

Decode

Decode speed

413.12 tok/s

Prefill

Prefill speed

1.03 s

TTFT

Time to first token

15.2 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

11 days ago

Last verified

Performance

  1. vLLM · Q4_K_S (current)Decode 94.16 · Prefill 413.12 ·
  2. Strata · Q3_K_S Decode 133.25 · Prefill 628.12 ·

Core figures

Decode (avg)
94.16 tok/s
Prefill (avg)
413.12 tok/s
TTFT (avg)
1.03 s
VRAM (avg)
15.2 GB
MTP acceptance rate
—
TTFB
0.75 GB
Power draw
304.2 W

Configuration

Member-level fields are taken from the most recent run

Model
Ministral-3-8B-Instruct-2512
Quantization
Q4_K_S
Framework
vLLM
Version
v0.34
Context length
16384 tokens
Batch size
8
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 3090
Nominal VRAM
24 GB
Measured VRAM (avg)
15.2 GB
OS
Windows 11 24H2
Driver
572.83
CUDA
12.4
Power draw
304.2 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-24

    v0.34 · Windows 11 24H2 · CUDA 12.4 · 16384 ctx

    94.16 tok/s

    Decode

    413.12 tok/s

    Prefill

    1.03 s

    TTFT

    15.2 GB

    VRAM

    —

    MTP

    304.2 W

    Power draw

    L2 batch seed 2026-10-24: vLLM on hw_id=2 running model_id=74 at Q4_K_S