Arao AI
Back to search results

iFlytek Spark-X2.5-4B · IQ1_S

Start a discussion
AMD Ryzen AI Max+ 395128GBllama.cpp v0.50L2 Cross-framework verified

This page aggregates 1 real-world runs of iFlytek Spark-X2.5-4B (IQ1_S) on AMD Ryzen AI Max+ 395 with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

115.1 tok/s

Decode

Decode speed

632.07 tok/s

Prefill

Prefill speed

1.19 s

TTFT

Time to first token

5.63 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

12 days ago

Last verified

Performance

  1. llama.cpp · IQ1_S (current)Decode 115.1 · Prefill 632.07 ·
  2. vLLM · IQ4_XS Decode 84.12 · Prefill 383.03 ·

Core figures

Decode (avg)
115.1 tok/s
Prefill (avg)
632.07 tok/s
TTFT (avg)
1.19 s
VRAM (avg)
5.63 GB
MTP acceptance rate
—
TTFB
0.92 GB
Power draw
54.2 W

Configuration

Member-level fields are taken from the most recent run

Model
iFlytek Spark-X2.5-4B
Quantization
IQ1_S
Framework
llama.cpp
Version
v0.50
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 Ryzen AI Max+ 395
Nominal VRAM
128 GB
Measured VRAM (avg)
5.63 GB
OS
Ubuntu 24.04
Driver
amdgpu-pro 24.20
CUDA
12.8
Power draw
54.2 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-24

    v0.50 · Ubuntu 24.04 · CUDA 12.8 · 32768 ctx

    115.1 tok/s

    Decode

    632.07 tok/s

    Prefill

    1.19 s

    TTFT

    5.63 GB

    VRAM

    —

    MTP

    54.2 W

    Power draw

    L2 batch seed 2026-10-24: llama.cpp on hw_id=24 running model_id=75 at IQ1_S