Arao AI
Back to search results

Qwen3-8B · IQ3_S

Start a discussion
NVIDIA RTX 3060 12GB12GBOllama v0.42L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3-8B (IQ3_S) on NVIDIA RTX 3060 12GB with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

58.42 tok/s

Decode

Decode speed

197.59 tok/s

Prefill

Prefill speed

1.92 s

TTFT

Time to first token

11.4 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

17 days ago

Last verified

Performance

  1. vLLM · Q3_K_S Decode 73.32 · Prefill 238.24 ·
  2. Ollama · IQ3_S (current)Decode 58.42 · Prefill 197.59 ·
  3. llama.cpp · IQ2_M Decode 70.85 · Prefill 298.67 ·
  4. vLLM · IQ3_S Decode 64 · Prefill 232.21 ·
  5. llama.cpp · IQ3_S Decode 81.43 · Prefill 366.11 ·

Core figures

Decode (avg)
58.42 tok/s
Prefill (avg)
197.59 tok/s
TTFT (avg)
1.92 s
VRAM (avg)
11.4 GB
MTP acceptance rate
—
TTFB
0.47 GB
Power draw
141.6 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3-8B
Quantization
IQ3_S
Framework
Ollama
Version
v0.42
Context length
4096 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 3060 12GB
Nominal VRAM
12 GB
Measured VRAM (avg)
11.4 GB
OS
Ubuntu 22.04
Driver
572.83
CUDA
12.8
Power draw
141.6 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

    v0.42 · Ubuntu 22.04 · CUDA 12.8 · 4096 ctx

    58.42 tok/s

    Decode

    197.59 tok/s

    Prefill

    1.92 s

    TTFT

    11.4 GB

    VRAM

    —

    MTP

    141.6 W

    Power draw

    L2 batch seed 2026-10-19: Ollama on hw_id=15 running model_id=53 at IQ3_S