DeepSeek V4-Flash Completes Tasks at 1/105th the Cost of Competitors

mustafamhus · x · 2026-08-02

Mustafa Suleyman highlights that the core metric of AI economics is shifting from 'cost per token' to 'cost per completed task'. Citing data from Artificial Analysis, he notes that DeepSeek V4-Flash completes the same benchmark tasks as Fable at 1/105th the cost ($1 vs $105).

Related event: DeepSeek V4-Flash Costs 105x Less, But Stability and Benchmark Overfitting Questioned(17 posts)→

Original post →

More from Venture

Venture channel →