Arao AI
Back to search results

Qwen3.5-0.8B · IQ3_XS

Start a discussion
NVIDIA RTX 4080 Super16GBOllama v0.49L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.5-0.8B (IQ3_XS) on NVIDIA RTX 4080 Super with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

393.56 tok/s

Decode

Decode speed

2,131.15 tok/s

Prefill

Prefill speed

1.32 s

TTFT

Time to first token

3.14 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

17 days ago

Last verified

Performance

  1. vLLM · Q3_K_S Decode 302.38 · Prefill 993.94 ·
  2. vLLM · Q4_K_S Decode 370.71 · Prefill 1429.25 ·
  3. Ollama · IQ3_S Decode 352.12 · Prefill 1849.99 ·
  4. Ollama · IQ3_XS (current)Decode 393.56 · Prefill 2131.15 ·
  5. llama.cpp · Q4_K_M Decode 344.73 · Prefill 1737.55 ·
  6. vLLM · IQ1_S Decode 314.79 · Prefill 1409.84 ·

Core figures

Decode (avg)
393.56 tok/s
Prefill (avg)
2,131.15 tok/s
TTFT (avg)
1.32 s
VRAM (avg)
3.14 GB
MTP acceptance rate
—
TTFB
3.15 GB
Power draw
260.9 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.5-0.8B
Quantization
IQ3_XS
Framework
Ollama
Version
v0.49
Context length
8192 tokens
Batch size
512
GPU layers
28
Flash Attention
Off

Hardware

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

GPU
NVIDIA RTX 4080 Super
Nominal VRAM
16 GB
Measured VRAM (avg)
3.14 GB
OS
Windows 11 23H2
Driver
570.86
CUDA
12.4
Power draw
260.9 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-19

    v0.49 · Windows 11 23H2 · CUDA 12.4 · 8192 ctx

    393.56 tok/s

    Decode

    2,131.15 tok/s

    Prefill

    1.32 s

    TTFT

    3.14 GB

    VRAM

    —

    MTP

    260.9 W

    Power draw

    L2 batch seed 2026-10-19: Ollama on hw_id=20 running model_id=51 at IQ3_XS