DeepSeek-V4-Flash Hits Pareto Frontier With $0.03 Query Cost

DeepSeek-V4-Flash achieves 50-point intelligence index with $0.03 average cost, entering the Pareto Frontier for unmatched efficiency balance.

Woofun AI data shows that DeepSeek-V4-Flash has attained an intelligence index of approximately 50 in the latest Artificial Analysis evaluation. Based on official API metrics and a 99% cache hit rate, the average cost per task stands at roughly $0.03.

The model has entered the "Pareto Frontier" of intelligence and cost, meaning few tested models offer superior performance at a lower price. While higher capability typically drives up costs and lower prices often reduce performance, DeepSeek-V4-Flash currently maintains the most efficient balance between these two factors.