Arao AI
Back to search results

Kimi-K3 · IQ4_XS

Start a discussion
AMD Radeon RX 9070 XT16GBvLLM v0.34L2 Cross-framework verified

This page aggregates 1 real-world runs of Kimi-K3 (IQ4_XS) on AMD Radeon RX 9070 XT with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

107.13 tok/s

Decode

Decode speed

359.17 tok/s

Prefill

Prefill speed

0.27 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 · IQ3_XS Decode 118.99 · Prefill 614.09 ·
  2. llama.cpp · Q5_K_M Decode 84.68 · Prefill 344.07 ·
  3. vLLM · IQ4_XS (current)Decode 107.13 · Prefill 359.17 ·

Core figures

Decode (avg)
107.13 tok/s
Prefill (avg)
359.17 tok/s
TTFT (avg)
0.27 s
VRAM (avg)
1.05 GB
MTP acceptance rate
—
TTFB
0.86 GB
Power draw
269.9 W

Configuration

Member-level fields are taken from the most recent run

Model
Kimi-K3
Quantization
IQ4_XS
Framework
vLLM
Version
v0.34
Context length
16384 tokens
Batch size
8
GPU layers
999
Flash Attention
On

Hardware

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

GPU
AMD Radeon RX 9070 XT
Nominal VRAM
16 GB
Measured VRAM (avg)
1.05 GB
OS
Ubuntu 22.04
Driver
amdgpu 6.11
CUDA
12.4
Power draw
269.9 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

    v0.34 · Ubuntu 22.04 · CUDA 12.4 · 16384 ctx

    107.13 tok/s

    Decode

    359.17 tok/s

    Prefill

    0.27 s

    TTFT

    1.05 GB

    VRAM

    —

    MTP

    269.9 W

    Power draw

    L2 batch seed 2026-10-18: vLLM on hw_id=16 running model_id=73 at IQ4_XS