Writer introduces new AI model and upgraded harness to contain token costs
Writer has launched its new flagship model Palmyra X6, along with upgrades to its agentic harness, aiming to cut customer costs by up to 50%.
Writer, a company providing AI tools and agents for marketers, announced on Thursday the release of its new flagship model, Palmyra X6. The model is a post-training adaptation of Z.ai's open-source GLM-5.2, designed to offer ready-to-use capabilities at a significantly lower price.
According to the company, the new model, combined with changes to its harness infrastructure, could reduce costs for customers by as much as 50% on basic tasks. The company also unveiled substantial upgrades to its standard agentic harness. Both features became available to Writer clients on Thursday.
CEO May Habib said that enterprises are tired of constantly chasing the latest benchmarks and want to see costs flatten out. The new approach places particular emphasis on complex, multi-step tasks, which can be executed faster and with fewer tokens.
A recent paper by Writer researchers supports this strategy, showing that adjustments to harness efficiency often reduce costs more reliably than switching models. In their tests, costs fell by an average of 40%. The researchers noted that the harness is a component whose efficiency compounds across every model an organization runs, both present and future.
For customers, the experience remains model-agnostic: Palmyra X6 will work alongside other Writer models or external models imported via Azure or Amazon Bedrock. Habib also linked the drive for cost cutting to a growing distrust of major AI labs, which have a financial incentive to encourage higher token usage. She added that these labs do not fully understand how to help enterprises derive value from AI.
