Arao AI
Back to search results

DeepSeek-R1-Distill-Qwen-14B · Q3_K_L

Start a discussion
NVIDIA RTX 409024GBOllama v0.22L2 Cross-framework verified

This page aggregates 1 real-world runs of DeepSeek-R1-Distill-Qwen-14B (Q3_K_L) on NVIDIA RTX 4090 with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

117.29 tok/s

Decode

Decode speed

610.83 tok/s

Prefill

Prefill speed

2.21 s

TTFT

Time to first token

22.8 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. vLLM · IQ2_M Decode 163.66 · Prefill 743.62 ·
  2. Ollama · Q3_K_L (current)Decode 117.29 · Prefill 610.83 ·
  3. vLLM · Q5_K_M Decode 82.4 · Prefill 618 ·
  4. vLLM · Q4_K_M Decode 100.8 · Prefill 756 ·
  5. llama.cpp · Q5_K_M Decode 75.6 · Prefill 567 ·
  6. llama.cpp · Q4_K_M Decode 92.5 · Prefill 693.8 ·

Core figures

Decode (avg)
117.29 tok/s
Prefill (avg)
610.83 tok/s
TTFT (avg)
2.21 s
VRAM (avg)
22.8 GB
MTP acceptance rate
—
TTFB
0.94 GB
Power draw
339.9 W

Configuration

Member-level fields are taken from the most recent run

Model
DeepSeek-R1-Distill-Qwen-14B
Quantization
Q3_K_L
Framework
Ollama
Version
v0.22
Context length
32768 tokens
Batch size
256
GPU layers
20
Flash Attention
On

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)
22.8 GB
OS
Windows 11 24H2
Driver
565.57
CUDA
12.6
Power draw
339.9 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-18

    v0.22 · Windows 11 24H2 · CUDA 12.6 · 32768 ctx

    117.29 tok/s

    Decode

    610.83 tok/s

    Prefill

    2.21 s

    TTFT

    22.8 GB

    VRAM

    —

    MTP

    339.9 W

    Power draw

    L2 batch seed 2026-10-18: Ollama on hw_id=1 running model_id=71 at Q3_K_L