Arao AI
Back to search results

Ministral-3-8B-Instruct-2512 · Q3_K_S

Start a discussion
NVIDIA RTX 409024GBOllama v0.42L2 Cross-framework verified

This page aggregates 1 real-world runs of Ministral-3-8B-Instruct-2512 (Q3_K_S) on NVIDIA RTX 4090 with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

142.08 tok/s

Decode

Decode speed

478.74 tok/s

Prefill

Prefill speed

0.52 s

TTFT

Time to first token

13.02 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

11 days ago

Last verified

Performance

  1. Ollama · Q3_K_S (current)Decode 142.08 · Prefill 478.74 ·
  2. Strata · IQ2_M Decode 230.86 · Prefill 912.94 ·

Core figures

Decode (avg)
142.08 tok/s
Prefill (avg)
478.74 tok/s
TTFT (avg)
0.52 s
VRAM (avg)
13.02 GB
MTP acceptance rate
—
TTFB
1.14 GB
Power draw
396.1 W

Configuration

Member-level fields are taken from the most recent run

Model
Ministral-3-8B-Instruct-2512
Quantization
Q3_K_S
Framework
Ollama
Version
v0.42
Context length
4096 tokens
Batch size
256
GPU layers
33
Flash Attention
On

Hardware

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

GPU
NVIDIA RTX 4090
Nominal VRAM
24 GB
Measured VRAM (avg)
13.02 GB
OS
Windows 11 24H2
Driver
572.83
CUDA
12.8
Power draw
396.1 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

    v0.42 · Windows 11 24H2 · CUDA 12.8 · 4096 ctx

    142.08 tok/s

    Decode

    478.74 tok/s

    Prefill

    0.52 s

    TTFT

    13.02 GB

    VRAM

    —

    MTP

    396.1 W

    Power draw

    L2 batch seed 2026-10-24: Ollama on hw_id=1 running model_id=74 at Q3_K_S