Arao AI
Back to search results

Kimi-K3 · Q3_K_L

Start a discussion
AMD Radeon RX 7900 XTX24GBllama.cpp v0.35L2 Cross-framework verified

This page aggregates 1 real-world runs of Kimi-K3 (Q3_K_L) on AMD Radeon RX 7900 XTX with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

75 tok/s

Decode

Decode speed

318.12 tok/s

Prefill

Prefill speed

1.01 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 · IQ1_S Decode 109.82 · Prefill 547.49 ·
  2. llama.cpp · Q3_K_L (current)Decode 75 · Prefill 318.12 ·
  3. Ollama · Q5_K_S Decode 56.93 · Prefill 285.15 ·

Core figures

Decode (avg)
75 tok/s
Prefill (avg)
318.12 tok/s
TTFT (avg)
1.01 s
VRAM (avg)
1.05 GB
MTP acceptance rate
—
TTFB
0.6 GB
Power draw
231.9 W

Configuration

Member-level fields are taken from the most recent run

Model
Kimi-K3
Quantization
Q3_K_L
Framework
llama.cpp
Version
v0.35
Context length
32768 tokens
Batch size
512
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
AMD Radeon RX 7900 XTX
Nominal VRAM
24 GB
Measured VRAM (avg)
1.05 GB
OS
Ubuntu 22.04
Driver
amdgpu 6.12
CUDA
12.4
Power draw
231.9 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

    v0.35 · Ubuntu 22.04 · CUDA 12.4 · 32768 ctx

    75 tok/s

    Decode

    318.12 tok/s

    Prefill

    1.01 s

    TTFT

    1.05 GB

    VRAM

    —

    MTP

    231.9 W

    Power draw

    L2 batch seed 2026-10-18: llama.cpp on hw_id=11 running model_id=73 at Q3_K_L