Arao AI
Back to search results

Qwen3.5-0.8B · IQ1_S

Start a discussion
NVIDIA RTX 409024GBvLLM v0.20L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.5-0.8B (IQ1_S) on NVIDIA RTX 4090 with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

328.9 tok/s

Decode

Decode speed

1,532.17 tok/s

Prefill

Prefill speed

0.53 s

TTFT

Time to first token

2.8 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. llama.cpp · IQ1_S Decode 373.87 · Prefill 1737.06 ·
  2. llama.cpp · Q3_K_S Decode 364.6 · Prefill 1679.71 ·
  3. vLLM · FP16 Decode 161.03 · Prefill 685.23 ·
  4. vLLM · IQ1_S (current)Decode 328.9 · Prefill 1532.17 ·
  5. llama.cpp · IQ2_XS Decode 280.41 · Prefill 1104.34 ·
  6. llama.cpp · Q4_K_M Decode 278.24 · Prefill 916.88 ·
  7. vLLM · IQ3_XS Decode 302.65 · Prefill 1148.6 ·

Core figures

Decode (avg)
328.9 tok/s
Prefill (avg)
1,532.17 tok/s
TTFT (avg)
0.53 s
VRAM (avg)
2.8 GB
MTP acceptance rate
—
TTFB
2.63 GB
Power draw
384.1 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.5-0.8B
Quantization
IQ1_S
Framework
vLLM
Version
v0.20
Context length
4096 tokens
Batch size
16
GPU layers
999
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)
2.8 GB
OS
Windows 11 23H2
Driver
565.57
CUDA
12.4
Power draw
384.1 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

    v0.20 · Windows 11 23H2 · CUDA 12.4 · 4096 ctx

    328.9 tok/s

    Decode

    1,532.17 tok/s

    Prefill

    0.53 s

    TTFT

    2.8 GB

    VRAM

    —

    MTP

    384.1 W

    Power draw

    L2 batch seed 2026-10-18: vLLM on hw_id=1 running model_id=51 at IQ1_S