DeepSeek to Reduce Flash Series API Prices Tomorrow
- DeepSeek announced price reductions for Flash series models effective September 10 at 12:00 Beijing time.
- Cache hit prices will decrease by a maximum of 60%.
- Cache miss prices will decrease by a maximum of 33.33%.
- Output prices will decrease by a maximum of 11.11%.
DeepSeek announced on September 9 that it will reduce pricing for its Flash series API models starting September 10 at 12:00 Beijing time. The price reductions vary by cost component: cache hits will see maximum reductions of 60%, cache misses 33.33%, and output tokens 11.11%. These changes apply across the Flash model line on the DeepSeek Open Platform.
For the cloud AI market, lower API pricing for a competitive model is price-positive for end users and may increase adoption pressure on other providers to adjust rates. The impact is neutral to slightly positive for token consumption (higher volume may offset lower margins) and depends on whether competitors respond with matching cuts.
DeepSeek将于明日下调Flash系列API价格
DeepSeek在9月9日宣布,将自9月10日北京时间12:00起下调Flash系列模型API价格。价格下调幅度按成本要素分别为:缓存命中价格最高下调60%,缓存未命中价格最高下调33.33%,输出token价格最高下调11.11%。这些调整适用于DeepSeek开放平台上的全系Flash模型。
对云AI市场而言,具有竞争力的模型价格下降对终端用户利好,可能增加其他供应商的价格调整压力。对token消耗的影响取决于交易量增加是否能抵消利润率下降,以及竞争对手是否响应式降价。