Arao AI
Back to search results

Qwen3.5-4B · Q3_K_S

Start a discussion
NVIDIA RTX 507012GBOllama v0.38L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.5-4B (Q3_K_S) on NVIDIA RTX 5070 with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

115.6 tok/s

Decode

Decode speed

548.24 tok/s

Prefill

Prefill speed

1.06 s

TTFT

Time to first token

7.56 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 (current)Decode 115.6 · Prefill 548.24 ·
  2. Ollama · Q5_K_S Decode 83.31 · Prefill 418.84 ·
  3. vLLM · Q3_K_M Decode 112.72 · Prefill 436.5 ·
  4. vLLM · Q4_K_M Decode 84.51 · Prefill 408.98 ·
  5. Strata · IQ3_S Decode 154.26 · Prefill 563.04 ·
  6. llama.cpp · Q5_K_M Decode 174.7 · Prefill 1310.3 ·

Core figures

Decode (avg)
115.6 tok/s
Prefill (avg)
548.24 tok/s
TTFT (avg)
1.06 s
VRAM (avg)
7.56 GB
MTP acceptance rate
—
TTFB
0.92 GB
Power draw
201.7 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.5-4B
Quantization
Q3_K_S
Framework
Ollama
Version
v0.38
Context length
32768 tokens
Batch size
512
GPU layers
20
Flash Attention
On

Hardware

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

GPU
NVIDIA RTX 5070
Nominal VRAM
12 GB
Measured VRAM (avg)
7.56 GB
OS
Ubuntu 24.04
Driver
572.83
CUDA
12.4
Power draw
201.7 W

Sources & evidence

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

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

    v0.38 · Ubuntu 24.04 · CUDA 12.4 · 32768 ctx

    115.6 tok/s

    Decode

    548.24 tok/s

    Prefill

    1.06 s

    TTFT

    7.56 GB

    VRAM

    —

    MTP

    201.7 W

    Power draw

    L2 batch seed 2026-10-18: Ollama on hw_id=18 running model_id=48 at Q3_K_S