Arao AI
Back to search results

Qwen3-8B · Q6_K

Start a discussion
NVIDIA RTX 409024GBllama.cpp v0.40L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3-8B (Q6_K) on NVIDIA RTX 4090 with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

88.33 tok/s

Decode

Decode speed

428.17 tok/s

Prefill

Prefill speed

1.75 s

TTFT

Time to first token

18.9 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. llama.cpp · Q6_K (current)Decode 88.33 · Prefill 428.17 ·
  2. llama.cpp · Q8_0 Decode 89.31 · Prefill 494.75 ·
  3. vLLM · Q4_K_S Decode 126.59 · Prefill 467.91 ·
  4. vLLM · Q8_0 Decode 76.65 · Prefill 513.21 ·
  5. Ollama · Q6_K Decode 82.66 · Prefill 332.47 ·
  6. llama.cpp · IQ4_XS Decode 112 · Prefill 1323 ·
  7. vLLM · IQ4_XS Decode 104 · Prefill 1176 ·
  8. vLLM · Q4_K_M Decode 96 · Prefill 1061 ·
  9. vLLM · Q5_K_M Decode 95 · Prefill 815 ·
  10. Ollama · Q4_K_M Decode 113 · Prefill 922 ·
  11. Ollama · Q5_K_M Decode 95 · Prefill 1090 ·
  12. llama.cpp · Q5_K_M Decode 133.3 · Prefill 999.8 ·
  13. llama.cpp · Q4_K_M Decode 160.5 · Prefill 1203.8 ·

Core figures

Decode (avg)
88.33 tok/s
Prefill (avg)
428.17 tok/s
TTFT (avg)
1.75 s
VRAM (avg)
18.9 GB
MTP acceptance rate
—
TTFB
0.71 GB
Power draw
393.4 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3-8B
Quantization
Q6_K
Framework
llama.cpp
Version
v0.40
Context length
8192 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)
18.9 GB
OS
macOS 15.0
Driver
565.57
CUDA
12.4
Power draw
393.4 W

Sources & evidence

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

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

    v0.40 · macOS 15.0 · CUDA 12.4 · 8192 ctx

    88.33 tok/s

    Decode

    428.17 tok/s

    Prefill

    1.75 s

    TTFT

    18.9 GB

    VRAM

    —

    MTP

    393.4 W

    Power draw

    L2 batch seed 2026-10-10: llama.cpp on hw_id=1 running model_id=53 at Q6_K