Arao AI
Back to search results

Qwen3.5-4B · Q3_K_L

Start a discussion
NVIDIA RTX 409024GBllama.cpp v0.37L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.5-4B (Q3_K_L) on NVIDIA RTX 4090 with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

174.81 tok/s

Decode

Decode speed

809.7 tok/s

Prefill

Prefill speed

1.2 s

TTFT

Time to first token

8.15 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. llama.cpp · Q3_K_L (current)Decode 174.81 · Prefill 809.7 ·
  2. vLLM · Q4_K_S Decode 181.53 · Prefill 809.47 ·
  3. Strata · Q6_K Decode 161.44 · Prefill 654.89 ·

Core figures

Decode (avg)
174.81 tok/s
Prefill (avg)
809.7 tok/s
TTFT (avg)
1.2 s
VRAM (avg)
8.15 GB
MTP acceptance rate
—
TTFB
1.4 GB
Power draw
330.4 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.5-4B
Quantization
Q3_K_L
Framework
llama.cpp
Version
v0.37
Context length
2048 tokens
Batch size
256
GPU layers
33
Flash Attention
Off

Hardware

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

GPU
NVIDIA RTX 4090
Nominal VRAM
24 GB
Measured VRAM (avg)
8.15 GB
OS
Ubuntu 22.04
Driver
560.94
CUDA
12.6
Power draw
330.4 W

Sources & evidence

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

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

    v0.37 · Ubuntu 22.04 · CUDA 12.6 · 2048 ctx

    174.81 tok/s

    Decode

    809.7 tok/s

    Prefill

    1.2 s

    TTFT

    8.15 GB

    VRAM

    —

    MTP

    330.4 W

    Power draw

    L2 batch seed 2026-10-10: llama.cpp on hw_id=1 running model_id=48 at Q3_K_L