Arao AI
Back to search results

Llama-3-8B-Instruct · Q2_K

Start a discussion
NVIDIA RTX 507012GBvLLM v0.47L2 Cross-framework verified

This page aggregates 1 real-world runs of Llama-3-8B-Instruct (Q2_K) on NVIDIA RTX 5070 with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

123.84 tok/s

Decode

Decode speed

585.88 tok/s

Prefill

Prefill speed

2.04 s

TTFT

Time to first token

11.34 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. vLLM · Q2_K (current)Decode 123.84 · Prefill 585.88 ·
  2. llama.cpp · Q8_0 Decode 60.9 · Prefill 456.8 ·
  3. llama.cpp · Q5_K_M Decode 88.8 · Prefill 666 ·
  4. llama.cpp · Q4_K_M Decode 107 · Prefill 802.5 ·

Core figures

Decode (avg)
123.84 tok/s
Prefill (avg)
585.88 tok/s
TTFT (avg)
2.04 s
VRAM (avg)
11.34 GB
MTP acceptance rate
—
TTFB
0.99 GB
Power draw
202.3 W

Configuration

Member-level fields are taken from the most recent run

Model
Llama-3-8B-Instruct
Quantization
Q2_K
Framework
vLLM
Version
v0.47
Context length
4096 tokens
Batch size
16
GPU layers
999
Flash Attention
On

Hardware

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

GPU
NVIDIA RTX 5070
Nominal VRAM
12 GB
Measured VRAM (avg)
11.34 GB
OS
Ubuntu 24.04
Driver
565.57
CUDA
12.4
Power draw
202.3 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

    v0.47 · Ubuntu 24.04 · CUDA 12.4 · 4096 ctx

    123.84 tok/s

    Decode

    585.88 tok/s

    Prefill

    2.04 s

    TTFT

    11.34 GB

    VRAM

    —

    MTP

    202.3 W

    Power draw

    L2 batch seed 2026-10-18: vLLM on hw_id=18 running model_id=69 at Q2_K