Arao AI
Back to search results

Llama-3-8B-Instruct · Q4_K_M

Start a discussion
NVIDIA RTX Spark128GBllama.cpp v0.32L2 Cross-framework verified

This page aggregates 1 real-world runs of Llama-3-8B-Instruct (Q4_K_M) on NVIDIA RTX Spark with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

50.11 tok/s

Decode

Decode speed

261.83 tok/s

Prefill

Prefill speed

1.93 s

TTFT

Time to first token

15.88 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

16 days ago

Last verified

Performance

  1. llama.cpp · Q4_K_M (current)Decode 50.11 · Prefill 261.83 ·
  2. llama.cpp · Q5_K_S Decode 49.33 · Prefill 259.05 ·
  3. Ollama · Q3_K_S Decode 56.77 · Prefill 268.38 ·
  4. Ollama · Q4_K_M Decode 51.27 · Prefill 167.27 ·

Core figures

Decode (avg)
50.11 tok/s
Prefill (avg)
261.83 tok/s
TTFT (avg)
1.93 s
VRAM (avg)
15.88 GB
MTP acceptance rate
—
TTFB
0.4 GB
Power draw
54.4 W

Configuration

Member-level fields are taken from the most recent run

Model
Llama-3-8B-Instruct
Quantization
Q4_K_M
Framework
llama.cpp
Version
v0.32
Context length
8192 tokens
Batch size
256
GPU layers
33
Flash Attention
On

Hardware

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

GPU
NVIDIA RTX Spark
Nominal VRAM
128 GB
Measured VRAM (avg)
15.88 GB
OS
Ubuntu 24.04
Driver
NVIDIA 590.10
CUDA
13.0
Power draw
54.4 W
Reference price
$2,399 · updated 2026-10-19

Prices are not live; check the store page

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-19

    v0.32 · Ubuntu 24.04 · CUDA 13.0 · 8192 ctx

    50.11 tok/s

    Decode

    261.83 tok/s

    Prefill

    1.93 s

    TTFT

    15.88 GB

    VRAM

    —

    MTP

    54.4 W

    Power draw

    L2 batch seed 2026-10-19: llama.cpp on hw_id=110 running model_id=69 at Q4_K_M