Arao AI
Back to search results

Qwen3.5-4B · IQ4_XS

Start a discussion
NVIDIA RTX 309024GBStrata v0.41L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.5-4B (IQ4_XS) on NVIDIA RTX 3090 with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.

157.12 tok/s

Decode

Decode speed

630.97 tok/s

Prefill

Prefill speed

1.5 s

TTFT

Time to first token

8.32 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. Ollama · Q3_K_S Decode 109.99 · Prefill 558.1 ·
  2. Ollama · Q8_0 Decode 62.23 · Prefill 337.34 ·
  3. Strata · IQ4_XS (current)Decode 157.12 · Prefill 630.97 ·
  4. llama.cpp · Q5_K_S Decode 139.06 · Prefill 548.16 ·
  5. llama.cpp · Q6_K Decode 108.41 · Prefill 420.36 ·
  6. vLLM · Q5_K_S Decode 129.61 · Prefill 696.02 ·

Core figures

Decode (avg)
157.12 tok/s
Prefill (avg)
630.97 tok/s
TTFT (avg)
1.5 s
VRAM (avg)
8.32 GB
MTP acceptance rate
0.74
TTFB
1.26 GB
Power draw
259.3 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.5-4B
Quantization
IQ4_XS
Framework
Strata
Version
v0.41
Context length
4096 tokens
Batch size
1024
GPU layers
8
Flash Attention
On

Hardware

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

GPU
NVIDIA RTX 3090
Nominal VRAM
24 GB
Measured VRAM (avg)
8.32 GB
OS
Windows 11 24H2
Driver
565.57
CUDA
13.0
Power draw
259.3 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-18

    v0.41 · Windows 11 24H2 · CUDA 13.0 · 4096 ctx

    157.12 tok/s

    Decode

    630.97 tok/s

    Prefill

    1.5 s

    TTFT

    8.32 GB

    VRAM

    0.74

    MTP

    259.3 W

    Power draw

    L2 batch seed 2026-10-18: Strata on hw_id=2 running model_id=48 at IQ4_XS