Arao AI
Back to search results

Kimi-K3 · Q4_K_M

Start a discussion
NVIDIA RTX 4070 Ti12GBllama.cpp v0.25L2 Cross-framework verified

This page aggregates 1 real-world runs of Kimi-K3 (Q4_K_M) on NVIDIA RTX 4070 Ti with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

95.6 tok/s

Decode

Decode speed

473.23 tok/s

Prefill

Prefill speed

1.45 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 · Q4_K_M (current)Decode 95.6 · Prefill 473.23 ·
  2. Ollama · Q3_K_S Decode 101.17 · Prefill 440.52 ·

Core figures

Decode (avg)
95.6 tok/s
Prefill (avg)
473.23 tok/s
TTFT (avg)
1.45 s
VRAM (avg)
1.05 GB
MTP acceptance rate
—
TTFB
0.76 GB
Power draw
264.1 W

Configuration

Member-level fields are taken from the most recent run

Model
Kimi-K3
Quantization
Q4_K_M
Framework
llama.cpp
Version
v0.25
Context length
8192 tokens
Batch size
512
GPU layers
28
Flash Attention
Off

Hardware

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

GPU
NVIDIA RTX 4070 Ti
Nominal VRAM
12 GB
Measured VRAM (avg)
1.05 GB
OS
Windows 11 24H2
Driver
570.86
CUDA
12.4
Power draw
264.1 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

    v0.25 · Windows 11 24H2 · CUDA 12.4 · 8192 ctx

    95.6 tok/s

    Decode

    473.23 tok/s

    Prefill

    1.45 s

    TTFT

    1.05 GB

    VRAM

    —

    MTP

    264.1 W

    Power draw

    L2 batch seed 2026-10-18: llama.cpp on hw_id=42 running model_id=73 at Q4_K_M