Arao AI
Back to search results

Llama-3.1-8B-Instruct · Q3_K_S

Start a discussion
Intel Arc B58012GBllama.cpp v0.48L2 Cross-framework verified

This page aggregates 1 real-world runs of Llama-3.1-8B-Instruct (Q3_K_S) on Intel Arc B580 with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

65.46 tok/s

Decode

Decode speed

319.04 tok/s

Prefill

Prefill speed

1.35 s

TTFT

Time to first token

11.4 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

16 days ago

Last verified

Performance

  1. llama.cpp · IQ3_S Decode 58.55 · Prefill 274.06 ·
  2. llama.cpp · Q3_K_S (current)Decode 65.46 · Prefill 319.04 ·
  3. Ollama · IQ2_XS Decode 70.97 · Prefill 264.22 ·
  4. Ollama · IQ2_M Decode 67.87 · Prefill 259.33 ·

Core figures

Decode (avg)
65.46 tok/s
Prefill (avg)
319.04 tok/s
TTFT (avg)
1.35 s
VRAM (avg)
11.4 GB
MTP acceptance rate
—
TTFB
0.52 GB
Power draw
164.8 W

Configuration

Member-level fields are taken from the most recent run

Model
Llama-3.1-8B-Instruct
Quantization
Q3_K_S
Framework
llama.cpp
Version
v0.48
Context length
4096 tokens
Batch size
256
GPU layers
33
Flash Attention
Off

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)
11.4 GB
OS
Ubuntu 24.04
Driver
Intel Arc 32.0.101.6552
CUDA
12.6
Power draw
164.8 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-19

    v0.48 · Ubuntu 24.04 · CUDA 12.6 · 4096 ctx

    65.46 tok/s

    Decode

    319.04 tok/s

    Prefill

    1.35 s

    TTFT

    11.4 GB

    VRAM

    —

    MTP

    164.8 W

    Power draw

    L2 batch seed 2026-10-19: llama.cpp on hw_id=94 running model_id=68 at Q3_K_S