DeepSeek launches V4.1-Flash, low-cost 552B MoE model

Chinese startup DeepSeek released V4.1-Flash, a 552-billion-parameter mixture-of-experts model with a new Causal Encoder-Decoder architecture activating only 8B-16B parameters. It claims superior performance to prior flagship models at substantially lower pricing and coincides with preparations for a Shanghai STAR Market IPO. The launch intensifies cost competition against frontier US and Chinese models.