On April 7th, Kimi Open Platform officially announced a price adjustment for its model inference services, significantly lowering the cost of context caching. This is based on a year's worth of technological advancements and performance optimization by Moonshot AI. This move signifies Kimi Open Platform's commitment to enhancing its technological capabilities, improving user experience, and promoting the widespread adoption of AI services.
According to Kimi Open Platform, this price adjustment is a result of Moonshot AI's significant breakthroughs in model training, inference acceleration, and resource utilization optimization over the past year. Through continuous technological iterations, the platform has not only improved model processing efficiency and performance but also reduced operating costs. The company stated that these achievements allow Kimi to pass on more savings to its users, further lowering the barrier to entry for businesses and developers accessing large language models.
The price adjustment generally lowers model inference service costs, with reductions varying depending on model specifications and usage scenarios. Simultaneously, context caching prices have also been significantly reduced. This is particularly noteworthy, as the Kimi platform is known for its ultra-long context processing capabilities. This adjustment will further improve cost-effectiveness for users handling long-text tasks.
Since its launch, Kimi Open Platform has positioned its support for ultra-long context input (up to 2 million characters) as its core competitive advantage, finding widespread application in scenarios such as document analysis, literature reviews, and code reproduction. The reduction in context caching prices means users will enjoy lower costs when handling complex tasks. For example, enterprise users or developers who frequently access large volumes of text data will see reduced expenses and improved development efficiency.
A developer who has been using the Kimi API for a long time commented, "The lower context caching price is a significant benefit for us. Previously, processing large documents was expensive; now we can accomplish more tasks with a smaller budget, significantly improving cost-effectiveness."