Arao AI
Back to search results

Llama-3.2-3B-Instruct · IQ2_XS

Start a discussion
NVIDIA H100 80GB HBM380GBOllama v0.41L2 Cross-framework verified

This page aggregates 1 real-world runs of Llama-3.2-3B-Instruct (IQ2_XS) on NVIDIA H100 80GB HBM3 with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

295.44 tok/s

Decode

Decode speed

1,300.71 tok/s

Prefill

Prefill speed

0.24 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 · FP16 Decode 379.95 · Prefill 1874.2 ·
  2. llama.cpp · Q5_K_M Decode 313.79 · Prefill 1508.71 ·
  3. Ollama · FP16 Decode 308.75 · Prefill 1127.72 ·
  4. Ollama · IQ2_XS (current)Decode 295.44 · Prefill 1300.71 ·

Core figures

Decode (avg)
295.44 tok/s
Prefill (avg)
1,300.71 tok/s
TTFT (avg)
0.24 s
VRAM (avg)
5.12 GB
MTP acceptance rate
—
TTFB
2.36 GB
Power draw
562.3 W

Configuration

Member-level fields are taken from the most recent run

Model
Llama-3.2-3B-Instruct
Quantization
IQ2_XS
Framework
Ollama
Version
v0.41
Context length
2048 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 H100 80GB HBM3
Nominal VRAM
80 GB
Measured VRAM (avg)
5.12 GB
OS
Ubuntu 24.04
Driver
570.86
CUDA
13.0
Power draw
562.3 W

Sources & evidence

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

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

    v0.41 · Ubuntu 24.04 · CUDA 13.0 · 2048 ctx

    295.44 tok/s

    Decode

    1,300.71 tok/s

    Prefill

    0.24 s

    TTFT

    5.12 GB

    VRAM

    —

    MTP

    562.3 W

    Power draw

    L2 batch seed 2026-10-19: Ollama on hw_id=55 running model_id=70 at IQ2_XS