According to CoinBeat Monitoring, Tencent Cloud AI Development Platform announced that starting from June 3, 2026, 00:00 (Beijing Time), it will significantly reduce the price of the DeepSeek-V4 series models, with the highest price cut reaching 97.5%, further reducing user's usage costs. This adjustment only involves price changes, and the model service capabilities remain unchanged.

Specifically, the inference input/output price of the DeepSeek-V4-Pro model has been reduced by 75% (adjusted to 1/4 of the original price), and the cache hit price has been slashed by up to 97.5%; the cache hit price of the DeepSeek-V4-Flash model has been reduced by 90% (adjusted to 1/10 of the original price). Tencent Cloud stated that it will continue to optimize the pricing strategy for large models and welcomes users to provide feedback through tickets.