Arao AI
Back to search results

Qwen3.5-4B · IQ2_M

Start a discussion
NVIDIA RTX 409024GBvLLM v0.53L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.5-4B (IQ2_M) on NVIDIA RTX 4090 with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

248.85 tok/s

Decode

Decode speed

1,225.63 tok/s

Prefill

Prefill speed

2.06 s

TTFT

Time to first token

6.47 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. llama.cpp · FP16 Decode 79.5 · Prefill 355.19 ·
  2. llama.cpp · IQ4_XS Decode 170.35 · Prefill 636.3 ·
  3. vLLM · IQ2_M (current)Decode 248.85 · Prefill 1225.63 ·
  4. vLLM · Q6_K Decode 133.61 · Prefill 629.89 ·
  5. llama.cpp · Q3_K_L Decode 174.81 · Prefill 809.7 ·
  6. vLLM · Q4_K_S Decode 181.53 · Prefill 809.47 ·
  7. Strata · Q6_K Decode 161.44 · Prefill 654.89 ·

Core figures

Decode (avg)
248.85 tok/s
Prefill (avg)
1,225.63 tok/s
TTFT (avg)
2.06 s
VRAM (avg)
6.47 GB
MTP acceptance rate
—
TTFB
1.99 GB
Power draw
331.8 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.5-4B
Quantization
IQ2_M
Framework
vLLM
Version
v0.53
Context length
2048 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)
6.47 GB
OS
Windows 11 24H2
Driver
560.94
CUDA
12.8
Power draw
331.8 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

    v0.53 · Windows 11 24H2 · CUDA 12.8 · 2048 ctx

    248.85 tok/s

    Decode

    1,225.63 tok/s

    Prefill

    2.06 s

    TTFT

    6.47 GB

    VRAM

    —

    MTP

    331.8 W

    Power draw

    L2 batch seed 2026-10-18: vLLM on hw_id=1 running model_id=48 at IQ2_M