Arao AI
Back to search results

iFlytek Spark-X2.5-4B · Q4_K_S

Start a discussion
NVIDIA RTX 3060 12GB12GBStrata v0.49L2 Cross-framework verified

This page aggregates 1 real-world runs of iFlytek Spark-X2.5-4B (Q4_K_S) on NVIDIA RTX 3060 12GB with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.

203.46 tok/s

Decode

Decode speed

1,004.82 tok/s

Prefill

Prefill speed

0.12 s

TTFT

Time to first token

8.65 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

11 days ago

Last verified

Performance

  1. Ollama · IQ2_M Decode 182.69 · Prefill 954.31 ·
  2. Strata · Q4_K_S (current)Decode 203.46 · Prefill 1004.82 ·

Core figures

Decode (avg)
203.46 tok/s
Prefill (avg)
1,004.82 tok/s
TTFT (avg)
0.12 s
VRAM (avg)
8.65 GB
MTP acceptance rate
0.74
TTFB
1.63 GB
Power draw
158.6 W

Configuration

Member-level fields are taken from the most recent run

Model
iFlytek Spark-X2.5-4B
Quantization
Q4_K_S
Framework
Strata
Version
v0.49
Context length
4096 tokens
Batch size
1024
GPU layers
5
Flash Attention
On

Hardware

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

GPU
NVIDIA RTX 3060 12GB
Nominal VRAM
12 GB
Measured VRAM (avg)
8.65 GB
OS
Windows 11 24H2
Driver
560.94
CUDA
12.8
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

    v0.49 · Windows 11 24H2 · CUDA 12.8 · 4096 ctx

    203.46 tok/s

    Decode

    1,004.82 tok/s

    Prefill

    0.12 s

    TTFT

    8.65 GB

    VRAM

    0.74

    MTP

    158.6 W

    Power draw

    L2 batch seed 2026-10-24: Strata on hw_id=15 running model_id=75 at Q4_K_S