DeepSeek: DeepSeek V4 Flash 0423

deepseek/deepseek-v4-flash

Context: 1,048,576In: $0.02/1MOut: $1.28/1MUpdated 2026-10-04

DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window. It is designed for fast inference and...

Intelligence
24.4
Coding
52.0
Agentic
26.3

Metric profile

Trend over time

Classic evals (curated from model cards)

DeepSeek — manually maintained, may lag releases

MMLU-Pro
75%
GPQA Diamond
67%
MATH-500
95%

Design Arena (ELO by category)

CategoryELOWin rateRank
website121449.1%
#51
codecategories121148.9%
#49
gamedev121050.2%
#46
3d120449.3%
#46
uicomponent117144.7%
#66
svg115548.4%
#40
dataviz113640.7%
#83
asciiart112242.8%
#55