option
Home
Flash News
Content
EdwardYoung
EdwardYoung
June 2, 2026

Tencent Cloud slashes DeepSeek V4 large model prices from June 3. Pro version inference costs drop 75% cache hit costs fall 97.5% while Flash model cache costs drop 90%. The MoE-based 1.6 trillion parameter models support 1M token context. Move aims to lower innovation barriers and intensify cloud AI deployment competition.

Tencent Cloud slashes DeepSeek V4 large model prices from June 3. Pro version inference costs drop 75% cache hit costs fall 97.5% while Flash model cache costs drop 90%. The MoE-based 1.6 trillion parameter models support 1M token context. Move aims to lower innovation barriers and intensify cloud AI deployment competition.
Comments (0)
0/300
OR