Arao AI
Back to search results

Llama-3.2-3B-Instruct · IQ2_XS

Start a discussion
NVIDIA RTX Spark128GBOllama v0.25L2 Cross-framework verified

This page aggregates 1 real-world runs of Llama-3.2-3B-Instruct (IQ2_XS) on NVIDIA RTX Spark with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

115.72 tok/s

Decode

Decode speed

413.02 tok/s

Prefill

Prefill speed

0.66 s

TTFT

Time to first token

5.12 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 · IQ3_S Decode 108.82 · Prefill 395.74 ·
  2. llama.cpp · Q4_K_M Decode 89.55 · Prefill 296 ·
  3. Ollama · IQ2_XS (current)Decode 115.72 · Prefill 413.02 ·
  4. Ollama · Q4_K_M Decode 76.59 · Prefill 359.8 ·

Core figures

Decode (avg)
115.72 tok/s
Prefill (avg)
413.02 tok/s
TTFT (avg)
0.66 s
VRAM (avg)
5.12 GB
MTP acceptance rate
—
TTFB
0.93 GB
Power draw
56.2 W

Configuration

Member-level fields are taken from the most recent run

Model
Llama-3.2-3B-Instruct
Quantization
IQ2_XS
Framework
Ollama
Version
v0.25
Context length
8192 tokens
Batch size
256
GPU layers
99
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)
5.12 GB
OS
Ubuntu 24.04
Driver
NVIDIA 590.10
CUDA
13.0
Power draw
56.2 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

    v0.25 · Ubuntu 24.04 · CUDA 13.0 · 8192 ctx

    115.72 tok/s

    Decode

    413.02 tok/s

    Prefill

    0.66 s

    TTFT

    5.12 GB

    VRAM

    —

    MTP

    56.2 W

    Power draw

    L2 batch seed 2026-10-19: Ollama on hw_id=110 running model_id=70 at IQ2_XS