Arao AI
Back to search results

DeepSeek-R1-Distill-Qwen-14B · IQ2_XS

Start a discussion
NVIDIA RTX 508016GBvLLM v0.31L2 Cross-framework verified

This page aggregates 1 real-world runs of DeepSeek-R1-Distill-Qwen-14B (IQ2_XS) on NVIDIA RTX 5080 with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

227.28 tok/s

Decode

Decode speed

1,043.49 tok/s

Prefill

Prefill speed

0.18 s

TTFT

Time to first token

15.2 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

16 days ago

Last verified

Performance

  1. vLLM · IQ2_XS (current)Decode 227.28 · Prefill 1043.49 ·
  2. llama.cpp · IQ2_XS Decode 202.82 · Prefill 808.1 ·

Core figures

Decode (avg)
227.28 tok/s
Prefill (avg)
1,043.49 tok/s
TTFT (avg)
0.18 s
VRAM (avg)
15.2 GB
MTP acceptance rate
—
TTFB
1.82 GB
Power draw
334.2 W

Configuration

Member-level fields are taken from the most recent run

Model
DeepSeek-R1-Distill-Qwen-14B
Quantization
IQ2_XS
Framework
vLLM
Version
v0.31
Context length
8192 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 5080
Nominal VRAM
16 GB
Measured VRAM (avg)
15.2 GB
OS
Windows 11 23H2
Driver
570.86
CUDA
12.4
Power draw
334.2 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

    v0.31 · Windows 11 23H2 · CUDA 12.4 · 8192 ctx

    227.28 tok/s

    Decode

    1,043.49 tok/s

    Prefill

    0.18 s

    TTFT

    15.2 GB

    VRAM

    —

    MTP

    334.2 W

    Power draw

    L2 batch seed 2026-10-19: vLLM on hw_id=14 running model_id=71 at IQ2_XS