As AI technology becomes more widespread in enterprises, high token costs have made many users feel overwhelmed. To help companies break out of the "the more you use, the more expensive it gets" dilemma, AI service provider Writer officially launched its new flagship model Palmyra X6 this Thursday and made significant upgrades to its underlying intelligent agent runtime program (harness).

According to reports, Palmyra X6 is built upon a post-training variant of the open-source model GLM-5.2 from Z.ai. With the new runtime framework optimization, the system can significantly reduce customer costs by up to 50% for basic tasks. The CEO of the company, May Habib, stated that the current industry situation has left many CIOs frustrated, and what enterprises need is not blindly chasing the next performance ceiling, but truly controllable costs.

In addition to launching the new model, the architectural optimization of the runtime program has also become a core highlight of this upgrade. A recent study released by the company shows that fine-tuning runtime efficiency is often a more reliable way to reduce costs compared to simply selecting different large models, with test results showing an average cost reduction of 40%. As a fundamental component, the runtime framework can significantly improve the efficiency of running all models within an organization.

Currently, Palmyra X6 is fully available to enterprise customers. Users can combine it with existing tools or call it through a cloud platform. Industry insiders believe that this move reflects changes in the current market environment: in the context of continuously rising AI deployment costs, how to control expenses through technological means has become the top priority for major enterprises.