Writer claims 50% cost cuts without switching models
writer released palmyra x6 and an upgraded harness to cut token costs by 50% for basic tasks. it signals a shift where enterprises optimize infrastructure rather than chasing new benchmarks to curb runaway ai bills.
writer released palmyra x6, a post-training variation of z.ai's glm-5.2, alongside an upgraded agentic harness to address rising deployment costs. the company claims the combined updates reduce token costs by 50% for basic tasks, supported by internal research showing harness optimization can lower expenses by 40% more effectively than model swapping alone. ceo may habib argues that enterprises are shifting focus from chasing benchmarks to stabilizing costs amid dissatisfaction with major labs. the system remains model-agnostic, allowing clients to run palmyra x6 alongside external models from azure or amazon bedrock.