Arao AI
Back to search results

Kimi-K3 · Q8_0

Start a discussion
NVIDIA RTX 409024GBOllama v0.54L2 Cross-framework verified

This page aggregates 1 real-world runs of Kimi-K3 (Q8_0) on NVIDIA RTX 4090 with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

70.56 tok/s

Decode

Decode speed

253.27 tok/s

Prefill

Prefill speed

0.65 s

TTFT

Time to first token

1.05 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. llama.cpp · IQ4_XS Decode 123.9 · Prefill 634.46 ·
  2. Ollama · Q3_K_L Decode 113.94 · Prefill 395.79 ·
  3. Ollama · Q8_0 (current)Decode 70.56 · Prefill 253.27 ·
  4. SGLang · Q5_K_M Decode — · Prefill — ·
  5. SGLang · Q4_K_M Decode — · Prefill — ·
  6. vLLM · Q5_K_M Decode — · Prefill — ·
  7. vLLM · Q4_K_M Decode — · Prefill — ·

Core figures

Decode (avg)
70.56 tok/s
Prefill (avg)
253.27 tok/s
TTFT (avg)
0.65 s
VRAM (avg)
1.05 GB
MTP acceptance rate
—
TTFB
0.56 GB
Power draw
325.9 W

Configuration

Member-level fields are taken from the most recent run

Model
Kimi-K3
Quantization
Q8_0
Framework
Ollama
Version
v0.54
Context length
4096 tokens
Batch size
256
GPU layers
33
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)
1.05 GB
OS
Windows 11 23H2
Driver
572.83
CUDA
12.6
Power draw
325.9 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

    v0.54 · Windows 11 23H2 · CUDA 12.6 · 4096 ctx

    70.56 tok/s

    Decode

    253.27 tok/s

    Prefill

    0.65 s

    TTFT

    1.05 GB

    VRAM

    —

    MTP

    325.9 W

    Power draw

    L2 batch seed 2026-10-18: Ollama on hw_id=1 running model_id=73 at Q8_0