Arao AI
Back to search results

iFlytek Spark-X2.5-4B · IQ2_M

Start a discussion
Intel Arc B58012GBllama.cpp v0.30L2 Cross-framework verified

This page aggregates 1 real-world runs of iFlytek Spark-X2.5-4B (IQ2_M) on Intel Arc B580 with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

164.15 tok/s

Decode

Decode speed

679.38 tok/s

Prefill

Prefill speed

1.9 s

TTFT

Time to first token

6.47 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

11 days ago

Last verified

Performance

  1. llama.cpp · IQ2_M (current)Decode 164.15 · Prefill 679.38 ·
  2. Ollama · Q3_K_L Decode 127.75 · Prefill 669.2 ·

Core figures

Decode (avg)
164.15 tok/s
Prefill (avg)
679.38 tok/s
TTFT (avg)
1.9 s
VRAM (avg)
6.47 GB
MTP acceptance rate
—
TTFB
1.31 GB
Power draw
158.6 W

Configuration

Member-level fields are taken from the most recent run

Model
iFlytek Spark-X2.5-4B
Quantization
IQ2_M
Framework
llama.cpp
Version
v0.30
Context length
65536 tokens
Batch size
256
GPU layers
99
Flash Attention
Off

Hardware

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

GPU
Intel Arc B580
Nominal VRAM
12 GB
Measured VRAM (avg)
6.47 GB
OS
Windows 11 24H2
Driver
Intel Arc 32.0.101.6552
CUDA
12.6
Power draw
158.6 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

    v0.30 · Windows 11 24H2 · CUDA 12.6 · 65536 ctx

    164.15 tok/s

    Decode

    679.38 tok/s

    Prefill

    1.9 s

    TTFT

    6.47 GB

    VRAM

    —

    MTP

    158.6 W

    Power draw

    L2 batch seed 2026-10-24: llama.cpp on hw_id=94 running model_id=75 at IQ2_M