Arao AI
Back to search results

Llama-3.2-3B-Instruct · IQ4_XS

Start a discussion
NVIDIA RTX 508016GBOllama v0.45L2 Cross-framework verified

This page aggregates 1 real-world runs of Llama-3.2-3B-Instruct (IQ4_XS) on NVIDIA RTX 5080 with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

362.6 tok/s

Decode

Decode speed

1,325.05 tok/s

Prefill

Prefill speed

2.25 s

TTFT

Time to first token

6.76 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. llama.cpp · IQ1_S Decode 351.28 · Prefill 1853.09 ·
  2. Ollama · IQ4_XS (current)Decode 362.6 · Prefill 1325.05 ·

Core figures

Decode (avg)
362.6 tok/s
Prefill (avg)
1,325.05 tok/s
TTFT (avg)
2.25 s
VRAM (avg)
6.76 GB
MTP acceptance rate
—
TTFB
2.9 GB
Power draw
204.5 W

Configuration

Member-level fields are taken from the most recent run

Model
Llama-3.2-3B-Instruct
Quantization
IQ4_XS
Framework
Ollama
Version
v0.45
Context length
2048 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
NVIDIA RTX 5080
Nominal VRAM
16 GB
Measured VRAM (avg)
6.76 GB
OS
Ubuntu 22.04
Driver
570.86
CUDA
12.8
Power draw
204.5 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-18

    v0.45 · Ubuntu 22.04 · CUDA 12.8 · 2048 ctx

    362.6 tok/s

    Decode

    1,325.05 tok/s

    Prefill

    2.25 s

    TTFT

    6.76 GB

    VRAM

    —

    MTP

    204.5 W

    Power draw

    L2 batch seed 2026-10-18: Ollama on hw_id=14 running model_id=70 at IQ4_XS