Arao AI
Back to search results

Llama-3.1-8B-Instruct · IQ2_M

Start a discussion
NVIDIA RTX 3060 12GB12GBvLLM v0.42L2 Cross-framework verified

This page aggregates 1 real-world runs of Llama-3.1-8B-Instruct (IQ2_M) on NVIDIA RTX 3060 12GB with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

92.78 tok/s

Decode

Decode speed

311.99 tok/s

Prefill

Prefill speed

0.25 s

TTFT

Time to first token

10.84 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. vLLM · IQ2_M (current)Decode 92.78 · Prefill 311.99 ·

Core figures

Decode (avg)
92.78 tok/s
Prefill (avg)
311.99 tok/s
TTFT (avg)
0.25 s
VRAM (avg)
10.84 GB
MTP acceptance rate
—
TTFB
0.74 GB
Power draw
150.7 W

Configuration

Member-level fields are taken from the most recent run

Model
Llama-3.1-8B-Instruct
Quantization
IQ2_M
Framework
vLLM
Version
v0.42
Context length
8192 tokens
Batch size
32
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)
10.84 GB
OS
Windows 11 24H2
Driver
572.83
CUDA
12.6
Power draw
150.7 W

Sources & evidence

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

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

    v0.42 · Windows 11 24H2 · CUDA 12.6 · 8192 ctx

    92.78 tok/s

    Decode

    311.99 tok/s

    Prefill

    0.25 s

    TTFT

    10.84 GB

    VRAM

    —

    MTP

    150.7 W

    Power draw

    L2 batch seed 2026-10-18: vLLM on hw_id=15 running model_id=68 at IQ2_M