Arao AI
Back to search results

Qwen3.5-4B · Q6_K

Start a discussion
NVIDIA RTX 408016GBStrata v0.43L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.5-4B (Q6_K) on NVIDIA RTX 4080 with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.

181.02 tok/s

Decode

Decode speed

665.46 tok/s

Prefill

Prefill speed

0.32 s

TTFT

Time to first token

10.5 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. Ollama · IQ2_XS Decode 194.9 · Prefill 903.61 ·
  2. Strata · Q6_K (current)Decode 181.02 · Prefill 665.46 ·
  3. vLLM · Q2_K Decode 151.45 · Prefill 529.44 ·
  4. vLLM · Q3_K_M Decode 130.67 · Prefill 528.13 ·
  5. Ollama · Q4_K_M Decode 95.12 · Prefill 401.62 ·

Core figures

Decode (avg)
181.02 tok/s
Prefill (avg)
665.46 tok/s
TTFT (avg)
0.32 s
VRAM (avg)
10.5 GB
MTP acceptance rate
0.65
TTFB
1.45 GB
Power draw
316.1 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.5-4B
Quantization
Q6_K
Framework
Strata
Version
v0.43
Context length
32768 tokens
Batch size
256
GPU layers
12
Flash Attention
On

Hardware

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

GPU
NVIDIA RTX 4080
Nominal VRAM
16 GB
Measured VRAM (avg)
10.5 GB
OS
Ubuntu 22.04
Driver
565.57
CUDA
12.4
Power draw
316.1 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

    v0.43 · Ubuntu 22.04 · CUDA 12.4 · 32768 ctx

    181.02 tok/s

    Decode

    665.46 tok/s

    Prefill

    0.32 s

    TTFT

    10.5 GB

    VRAM

    0.65

    MTP

    316.1 W

    Power draw

    L2 batch seed 2026-10-18: Strata on hw_id=41 running model_id=48 at Q6_K