Arao AI
Back to search results

iFlytek Spark-X2.5-4B · Q3_K_S

Start a discussion
Apple M5 Max128GBllama.cpp v0.38L2 Cross-framework verified

This page aggregates 1 real-world runs of iFlytek Spark-X2.5-4B (Q3_K_S) on Apple M5 Max with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

215.47 tok/s

Decode

Decode speed

840.6 tok/s

Prefill

Prefill speed

0.29 s

TTFT

Time to first token

7.56 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 · Q3_K_S (current)Decode 215.47 · Prefill 840.6 ·
  2. Ollama · Q2_K Decode 217.16 · Prefill 1089.96 ·

Core figures

Decode (avg)
215.47 tok/s
Prefill (avg)
840.6 tok/s
TTFT (avg)
0.29 s
VRAM (avg)
7.56 GB
MTP acceptance rate
—
TTFB
1.72 GB
Power draw
58.5 W

Configuration

Member-level fields are taken from the most recent run

Model
iFlytek Spark-X2.5-4B
Quantization
Q3_K_S
Framework
llama.cpp
Version
v0.38
Context length
16384 tokens
Batch size
512
GPU layers
99
Flash Attention
On

Hardware

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

GPU
Apple M5 Max
Nominal VRAM
128 GB
Measured VRAM (avg)
7.56 GB
OS
macOS 26.1
Power draw
58.5 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

    v0.38 · macOS 26.1 · 16384 ctx

    215.47 tok/s

    Decode

    840.6 tok/s

    Prefill

    0.29 s

    TTFT

    7.56 GB

    VRAM

    —

    MTP

    58.5 W

    Power draw

    L2 batch seed 2026-10-24: llama.cpp on hw_id=28 running model_id=75 at Q3_K_S