SAN FRANCISCO — August 13, 2026 — WRITER, the enterprise AI agent platform trusted by the world’s leading Fortune 500 brands, today released Palmyra X6, its new flagship model that brings frontier-competitive performance to the critical workflows marketing and revenue teams run every day work, at a fraction of the cost of leading models; major upgrades to the WRITER Agent harness that make complex, multi-step workflows more cost-efficient and reliable from start to finish; and new governance tools that help leaders control token spend and scale the strongest use cases across the business. Together, the upgrades allow marketing and revenue teams to confidently scale their most ambitious agentic workflows in production without token costs holding them back.
Agents have moved token economics to the center of the enterprise AI conversation. As agents move into production, cost per task can jump by orders of magnitude compared to traditional chat experiences. Agents are now reasoning through complex, multi-step workflows, calling tools, and executing longer-running work, with every action consuming more tokens, taking more time, and creating another point where the workflow can derail. WRITER Agent now operates at an average 52% lower cost with 48% improvement in speed and 10% improvement in quality when paired with Palmyra X6 – delivering top tier performance at a fraction of the cost.
“The enterprise wants token consumption to explode — it means adoption is happening — but they need costs to flatten,” said Waseem AlShikh, CTO and Co-Founder of WRITER. “For the past five years, WRITER has focused on delivering not just frontier-level model performance, but the efficiency, reliability, and governance that highly regulated enterprises need to confidently put agents into production. With Palmyra X6 and major harness improvements to WRITER Agent, we’re delivering a step change in both agent capabilities and cost that fundamentally changes how broadly and boldly our customers can deploy agentic AI.”
Palmyra X6 is WRITER’s new flagship model, engineered to keep cost and time per unit of work low, while delivering frontier-competitive performance. Built for the enterprise, the model was designed to run inside the WRITER Agent harness and excels at marketing and revenue workflows.
To develop X6 for the specific demands of enterprise work, WRITER created new internal evaluations based on production workflows its customers actually run. Unlike public benchmarks, which measure general model performance, WRITER’s evaluation tests the capabilities enterprise teams rely on every day, including whether the model stays grounded in company knowledge, understands context, follows guidelines and guardrails, uses enterprise tools and remains coherent across long-running tasks.
The evaluation measures nine capabilities, including grounding and retrieval, tool use, content generation, sub-agent delegation and brand voice. X6 was post-trained on top of GLM 5.2, the strongest available open-weight model, to excel across those capabilities inside WRITER Agent. X6 delivered the best combination of capability and cost among the models WRITER tested, scoring an average of 0.87 (out of 1.00) across all nine evaluations at $2 per million input tokens and $8 per million output tokens, ahead of Claude Opus 4.8 (0.86 at $15/$75 per million), Claude Sonnet 4.6 (0.85 at $3/$15), GPT-5.5 (0.80 at $5/$15) and Gemini 3.1 (0.77 at $2.50/$10). X6 also tested as the least politically biased model WRITER evaluated, a critical attribute for marketing and revenue teams that need confidence in the content they generate and publish under their company’s brand.
X6 completes tasks in 26 seconds on average, generates 82 tokens per second, and can work unattended toward a single goal for up to eight hours, sustaining coherent reasoning across long, multi-stage objectives without drifting off task.
The WRITER Agent harness has been rebuilt to execute complex, multi-step work with greater speed and reliability while using fewer tokens, regardless of the underlying model.
On the backend, WRITER Agent now dynamically adapts reasoning based on the task at hand, answering simple questions directly or developing structured plans for more complex work. It can also execute high-volume work faster by running tasks in batches or delegating to sub-agents, allowing agents to complete longer, more complex workflows without unnecessary steps or repeated context.
New research from WRITER, The Harness Effect: How Orchestration Design Sets the Token Economics of Enterprise Agentic AI, shows these harness upgrades improve performance regardless of the model powering the task. Across all WRITER and third-party models tested, WRITER Agent completed tasks 44% faster, at a 41% lower cost per task on average, while maintaining quality. Paired with Palmyra X6, WRITER Agent delivered the strongest combined performance across cost, quality, speed, and reliability.
With today’s release, WRITER’s multi-model support now extends to WRITER Agent, giving teams a selection of popular models to work with. Admins can enable models from providers like Anthropic and OpenAI, letting users across their managed teams choose a model at the start of a session. Specialized models can also be selected for specific capabilities, such as image generation. And beyond WRITER’s own catalog, admins can bring their own models from cloud providers including Microsoft Azure, AWS Bedrock, and NVIDIA NIM.
“The conversation in marketing has shifted from exploring AI use cases to discussing what it costs to run,” said Roger Beharry Lall, Research Director: Advertising Technologies and SMB Marketing Applications, IDC. “Agentic workflows are one area where costs are becoming a major concern. Every extra step and tool call adds tokens, and at scale that adds up fast. In fact, IDC research finds that three-quarters of organizations worldwide now point to excessive AI spending as a major risk to their plans. Training, governance, and workflow design remain important, but what matters now is efficient agentic execution of the work marketers do every day, at a price point that lets teams run agents broadly instead of saving them for a handful of high-value tasks.”
WRITER also launched a major upgrade to reporting and governance controls that gives leaders a clear view into the agentic workflows being used across their teams. As more employees build and share Playbooks and Skills – repeatable agent workflows and instructions – leaders can see which use cases are seeing the highest adoption, identify and scale the best Playbooks and Skills across teams, and stay ahead of the biggest drivers of token spend. New capabilities include:
With a clearer view into adoption, performance, and spend, leaders can invest in the winning workflows and scale the ways of working that deliver the strongest results across the business.
“Governance gave us the confidence to scale AI past a few pilots,” said Robyn Murphy, Chief of Staff & Director, Marketing Operations, H&R Block. “Now, with Palmyra X6 bringing the cost per task down, we will have more confidence to take on high-volume work that was never worth automating before. And with the tools to look at usage, we can better understand which of those workflows will drive transformation.”
All new features are available today for WRITER customers. Visit WRITER.com to learn more.
WRITER is the enterprise AI agent platform trusted by Fortune 500 companies globally. Built for high-stakes enterprise environments, WRITER gives marketing, sales, and business teams powerful AI teammates that can plan, execute, and scale on-brand work across company systems, with rich organizational context encoded directly in every output. WRITER offers IT teams full governance and granular control, while business teams gain the intuitive tools and support to drive efficiency, capacity, and revenue growth at unprecedented speed and scale. Founded in 2020, WRITER is delivering unmatched ROI for hundreds of customers and is backed by investors including Insight Partners, ICONIQ, Radical Ventures, Premji Invest, Balderton, B Capital, Salesforce Ventures, Adobe Ventures, Accenture, Citi Ventures, IBM Ventures, WndrCo, and others.
Learn more at WRITER.com.
Media Contact: [email protected]