Arao AI
Back to search results

Qwen3-4B · IQ2_XS

Start a discussion
NVIDIA RTX 308010GBOllama v0.32L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3-4B (IQ2_XS) on NVIDIA RTX 3080 with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

118.69 tok/s

Decode

Decode speed

621.83 tok/s

Prefill

Prefill speed

2.44 s

TTFT

Time to first token

6.13 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

5 days ago

Last verified

Performance

  1. llama.cpp · IQ2_XS Decode 131.47 · Prefill 451.18 ·
  2. llama.cpp · Q3_K_L Decode 99.23 · Prefill 538.67 ·
  3. Ollama · IQ2_XS (current)Decode 118.69 · Prefill 621.83 ·
  4. Ollama · Q2_K Decode 102.28 · Prefill 452.09 ·

Core figures

Decode (avg)
118.69 tok/s
Prefill (avg)
621.83 tok/s
TTFT (avg)
2.44 s
VRAM (avg)
6.13 GB
MTP acceptance rate
—
TTFB
0.95 GB
Power draw
251.9 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3-4B
Quantization
IQ2_XS
Framework
Ollama
Version
v0.32
Context length
16384 tokens
Batch size
256
GPU layers
99
Flash Attention
Off

Hardware

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

GPU
NVIDIA RTX 3080
Nominal VRAM
10 GB
Measured VRAM (avg)
6.13 GB
OS
Ubuntu 22.04
Driver
570.86
CUDA
12.8
Power draw
251.9 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-13

    v0.32 · Ubuntu 22.04 · CUDA 12.8 · 16384 ctx

    118.69 tok/s

    Decode

    621.83 tok/s

    Prefill

    2.44 s

    TTFT

    6.13 GB

    VRAM

    —

    MTP

    251.9 W

    Power draw

    L2 batch seed 2026-10-13: Ollama on hw_id=44 running model_id=54 at IQ2_XS