Arao AI
Back to search results

Kimi-K3 · Q5_K_M

Start a discussion
AMD Radeon RX 9070 XT16GBllama.cpp v0.21L2 Cross-framework verified

This page aggregates 1 real-world runs of Kimi-K3 (Q5_K_M) on AMD Radeon RX 9070 XT with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

84.68 tok/s

Decode

Decode speed

344.07 tok/s

Prefill

Prefill speed

2.26 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 (current)Decode 84.68 · Prefill 344.07 ·
  3. vLLM · IQ4_XS Decode 107.13 · Prefill 359.17 ·

Core figures

Decode (avg)
84.68 tok/s
Prefill (avg)
344.07 tok/s
TTFT (avg)
2.26 s
VRAM (avg)
1.05 GB
MTP acceptance rate
—
TTFB
0.68 GB
Power draw
209.2 W

Configuration

Member-level fields are taken from the most recent run

Model
Kimi-K3
Quantization
Q5_K_M
Framework
llama.cpp
Version
v0.21
Context length
32768 tokens
Batch size
256
GPU layers
20
Flash Attention
Off

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 24.04
Driver
amdgpu 6.12
CUDA
12.4
Power draw
209.2 W

Sources & evidence

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

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

    v0.21 · Ubuntu 24.04 · CUDA 12.4 · 32768 ctx

    84.68 tok/s

    Decode

    344.07 tok/s

    Prefill

    2.26 s

    TTFT

    1.05 GB

    VRAM

    —

    MTP

    209.2 W

    Power draw

    L2 batch seed 2026-10-18: llama.cpp on hw_id=16 running model_id=73 at Q5_K_M