Arao AI
Back to search results

Llama-3-8B-Instruct · Q3_K_M

Start a discussion
NVIDIA RTX 3060 12GB12GBStrata v0.37L2 Cross-framework verified

This page aggregates 1 real-world runs of Llama-3-8B-Instruct (Q3_K_M) on NVIDIA RTX 3060 12GB with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.

95.19 tok/s

Decode

Decode speed

317.89 tok/s

Prefill

Prefill speed

0.3 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

17 days ago

Last verified

Performance

  1. Strata · Q3_K_M (current)Decode 95.19 · Prefill 317.89 ·
  2. Strata · IQ1_S Decode 126.53 · Prefill 600.27 ·
  3. Strata · Q3_K_S Decode 98.81 · Prefill 492.63 ·

Core figures

Decode (avg)
95.19 tok/s
Prefill (avg)
317.89 tok/s
TTFT (avg)
0.3 s
VRAM (avg)
11.4 GB
MTP acceptance rate
0.65
TTFB
0.76 GB
Power draw
147.4 W

Configuration

Member-level fields are taken from the most recent run

Model
Llama-3-8B-Instruct
Quantization
Q3_K_M
Framework
Strata
Version
v0.37
Context length
2048 tokens
Batch size
512
GPU layers
8
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)
11.4 GB
OS
Ubuntu 22.04
Driver
570.86
CUDA
12.8
Power draw
147.4 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

    v0.37 · Ubuntu 22.04 · CUDA 12.8 · 2048 ctx

    95.19 tok/s

    Decode

    317.89 tok/s

    Prefill

    0.3 s

    TTFT

    11.4 GB

    VRAM

    0.65

    MTP

    147.4 W

    Power draw

    L2 batch seed 2026-10-19: Strata on hw_id=15 running model_id=69 at Q3_K_M