Arao AI
Back to search results

Ministral-3-8B-Instruct-2512 · Q2_K

Start a discussion
Intel Arc B58012GBOllama v0.36L2 Cross-framework verified

This page aggregates 1 real-world runs of Ministral-3-8B-Instruct-2512 (Q2_K) on Intel Arc B580 with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

60.73 tok/s

Decode

Decode speed

314.93 tok/s

Prefill

Prefill speed

1.51 s

TTFT

Time to first token

11.34 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 · Q2_K (current)Decode 60.73 · Prefill 314.93 ·

Core figures

Decode (avg)
60.73 tok/s
Prefill (avg)
314.93 tok/s
TTFT (avg)
1.51 s
VRAM (avg)
11.34 GB
MTP acceptance rate
—
TTFB
0.49 GB
Power draw
164.6 W

Configuration

Member-level fields are taken from the most recent run

Model
Ministral-3-8B-Instruct-2512
Quantization
Q2_K
Framework
Ollama
Version
v0.36
Context length
16384 tokens
Batch size
256
GPU layers
28
Flash Attention
On

Hardware

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

GPU
Intel Arc B580
Nominal VRAM
12 GB
Measured VRAM (avg)
11.34 GB
OS
Windows 11 24H2
Driver
Intel Arc 32.0.101.6552
CUDA
12.4
Power draw
164.6 W

Sources & evidence

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

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

    v0.36 · Windows 11 24H2 · CUDA 12.4 · 16384 ctx

    60.73 tok/s

    Decode

    314.93 tok/s

    Prefill

    1.51 s

    TTFT

    11.34 GB

    VRAM

    —

    MTP

    164.6 W

    Power draw

    L2 batch seed 2026-10-24: Ollama on hw_id=94 running model_id=74 at Q2_K