Arao AI
Back to search results

Llama-3.1-8B-Instruct · IQ2_XS

Start a discussion
NVIDIA RTX 3060 12GB12GBvLLM v0.48L2 Cross-framework verified

This page aggregates 1 real-world runs of Llama-3.1-8B-Instruct (IQ2_XS) on NVIDIA RTX 3060 12GB with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

94.63 tok/s

Decode

Decode speed

430.36 tok/s

Prefill

Prefill speed

1.45 s

TTFT

Time to first token

10.16 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

17 days ago

Last verified

Performance

  1. llama.cpp · IQ3_S Decode 72.47 · Prefill 300.81 ·
  2. vLLM · IQ2_XS (current)Decode 94.63 · Prefill 430.36 ·
  3. vLLM · IQ2_M Decode 92.78 · Prefill 311.99 ·

Core figures

Decode (avg)
94.63 tok/s
Prefill (avg)
430.36 tok/s
TTFT (avg)
1.45 s
VRAM (avg)
10.16 GB
MTP acceptance rate
—
TTFB
0.76 GB
Power draw
167.7 W

Configuration

Member-level fields are taken from the most recent run

Model
Llama-3.1-8B-Instruct
Quantization
IQ2_XS
Framework
vLLM
Version
v0.48
Context length
4096 tokens
Batch size
8
GPU layers
999
Flash Attention
Off

Hardware

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

GPU
NVIDIA RTX 3060 12GB
Nominal VRAM
12 GB
Measured VRAM (avg)
10.16 GB
OS
Ubuntu 22.04
Driver
570.86
CUDA
12.4
Power draw
167.7 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 22.04 · CUDA 12.4 · 4096 ctx

    94.63 tok/s

    Decode

    430.36 tok/s

    Prefill

    1.45 s

    TTFT

    10.16 GB

    VRAM

    —

    MTP

    167.7 W

    Power draw

    L2 batch seed 2026-10-19: vLLM on hw_id=15 running model_id=68 at IQ2_XS