Arao AI
Back to search results

Kimi-K3 · IQ2_XS

Start a discussion
NVIDIA RTX 4080 Super16GBllama.cpp v0.28L2 Cross-framework verified

This page aggregates 1 real-world runs of Kimi-K3 (IQ2_XS) on NVIDIA RTX 4080 Super with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

184.95 tok/s

Decode

Decode speed

620.72 tok/s

Prefill

Prefill speed

0.68 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 · IQ2_XS (current)Decode 184.95 · Prefill 620.72 ·
  2. Strata · IQ1_S Decode 255.29 · Prefill 866.27 ·

Core figures

Decode (avg)
184.95 tok/s
Prefill (avg)
620.72 tok/s
TTFT (avg)
0.68 s
VRAM (avg)
1.05 GB
MTP acceptance rate
—
TTFB
1.48 GB
Power draw
283.7 W

Configuration

Member-level fields are taken from the most recent run

Model
Kimi-K3
Quantization
IQ2_XS
Framework
llama.cpp
Version
v0.28
Context length
32768 tokens
Batch size
512
GPU layers
20
Flash Attention
On

Hardware

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

GPU
NVIDIA RTX 4080 Super
Nominal VRAM
16 GB
Measured VRAM (avg)
1.05 GB
OS
Windows 11 23H2
Driver
565.57
CUDA
12.8
Power draw
283.7 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

    v0.28 · Windows 11 23H2 · CUDA 12.8 · 32768 ctx

    184.95 tok/s

    Decode

    620.72 tok/s

    Prefill

    0.68 s

    TTFT

    1.05 GB

    VRAM

    —

    MTP

    283.7 W

    Power draw

    L2 batch seed 2026-10-18: llama.cpp on hw_id=20 running model_id=73 at IQ2_XS