Arao AI
Back to search results

Llama-3.2-3B-Instruct · Q3_K_S

Start a discussion
NVIDIA RTX PRO 6000 Blackwell 96GB96GBOllama v0.38L2 Cross-framework verified

This page aggregates 1 real-world runs of Llama-3.2-3B-Instruct (Q3_K_S) on NVIDIA RTX PRO 6000 Blackwell 96GB with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

364.08 tok/s

Decode

Decode speed

1,984.53 tok/s

Prefill

Prefill speed

0.18 s

TTFT

Time to first token

6.2 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_XS Decode 338.03 · Prefill 1168.59 ·
  2. llama.cpp · Q8_0 Decode 331.08 · Prefill 1764.65 ·
  3. Ollama · IQ3_XS Decode 283.4 · Prefill 1452.77 ·
  4. Ollama · Q3_K_S (current)Decode 364.08 · Prefill 1984.53 ·

Core figures

Decode (avg)
364.08 tok/s
Prefill (avg)
1,984.53 tok/s
TTFT (avg)
0.18 s
VRAM (avg)
6.2 GB
MTP acceptance rate
—
TTFB
2.91 GB
Power draw
355.7 W

Configuration

Member-level fields are taken from the most recent run

Model
Llama-3.2-3B-Instruct
Quantization
Q3_K_S
Framework
Ollama
Version
v0.38
Context length
4096 tokens
Batch size
256
GPU layers
28
Flash Attention
On

Hardware

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

GPU
NVIDIA RTX PRO 6000 Blackwell 96GB
Nominal VRAM
96 GB
Measured VRAM (avg)
6.2 GB
OS
Windows 11 23H2
Driver
565.57
CUDA
12.4
Power draw
355.7 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

    v0.38 · Windows 11 23H2 · CUDA 12.4 · 4096 ctx

    364.08 tok/s

    Decode

    1,984.53 tok/s

    Prefill

    0.18 s

    TTFT

    6.2 GB

    VRAM

    —

    MTP

    355.7 W

    Power draw

    L2 batch seed 2026-10-19: Ollama on hw_id=68 running model_id=70 at Q3_K_S