Writer unveils Palmyra X6 model and upgraded harness, promising up to 50% lower costs
Enterprise AI company Writer has launched a new flagship model called Palmyra X6 alongside an upgraded agentic harness, saying the combination could cut customer costs on basic tasks by as much as 50%. Both features became available to clients on Thursday.
Writer, which builds AI tools and agents for marketers, announced on Thursday a new flagship model called Palmyra X6. The model is a post-training variation built on Z.ai's open source model GLM-5.2, and Writer says it delivers deployment-ready capabilities at a significantly lower price than comparable options.
The company estimates that the new model, combined with changes to its agentic harness infrastructure, could cut customer costs on basic tasks by as much as 50%. Alongside the new model, Writer also released major upgrades to its standard agentic harness. Both features became available to Writer clients starting Thursday.
Focus on cutting costs
Writer CEO May Habib told TechCrunch that enterprises are tired of chasing the latest benchmarks and instead want flattened, predictable costs, something she said no provider has managed to deliver so far. The new approach focuses particularly on complex, multi-step tasks, aiming to execute them faster and with fewer tokens.
A recent paper by Writer researchers supports this strategy, finding that in many cases, adjustments to the harness proved a more reliable way to cut costs than switching models, with average cost reductions of 40% across their tests. The researchers noted that harness efficiency affects every model an organization uses, both now and in the future.
For Writer's clients, the setup remains model-agnostic: Palmyra X6 will run alongside other Writer models or external models imported through Azure or Amazon Bedrock.
Habib also pointed to an unprecedented rise in costs for customers, which she said is fueling growing distrust among chief information officers toward major AI labs. According to Habib, these labs have a financial incentive to drive up token usage and do not fully understand how to help enterprises get real value from AI.
