logoalt Hacker News

terhechte • today at 6:08 PM • 0 replies • view on HN

Not really. “Flash” generally means fast or cost-efficient, not specifically MoE (Mixture of Experts).

- DeepSeek V4 Flash: 284B / 13B active MoE

- DeepSeek V4 Pro: 1.6T / 49B active MoE

- DeepSeek V4.1 Flash: 552B / 8B input, 16B output MoE