Arao AI
Back to search results

Qwen3.5-4B · Q8_0

Start a discussion
NVIDIA RTX 309024GBOllama v0.46L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.5-4B (Q8_0) on NVIDIA RTX 3090 with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

62.23 tok/s

Decode

Decode speed

337.34 tok/s

Prefill

Prefill speed

1.35 s

TTFT

Time to first token

11.76 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 (current)Decode 62.23 · Prefill 337.34 ·
  3. Strata · IQ4_XS 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)
62.23 tok/s
Prefill (avg)
337.34 tok/s
TTFT (avg)
1.35 s
VRAM (avg)
11.76 GB
MTP acceptance rate
—
TTFB
0.5 GB
Power draw
275.3 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.5-4B
Quantization
Q8_0
Framework
Ollama
Version
v0.46
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 3090
Nominal VRAM
24 GB
Measured VRAM (avg)
11.76 GB
OS
Ubuntu 24.04
Driver
565.57
CUDA
12.8
Power draw
275.3 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

    v0.46 · Ubuntu 24.04 · CUDA 12.8 · 16384 ctx

    62.23 tok/s

    Decode

    337.34 tok/s

    Prefill

    1.35 s

    TTFT

    11.76 GB

    VRAM

    —

    MTP

    275.3 W

    Power draw

    L2 batch seed 2026-10-18: Ollama on hw_id=2 running model_id=48 at Q8_0