SAN FRANCISCO--(BUSINESS WIRE)--Writer, the full-stack generative AI platform for the enterprise, today released its newest large language model (LLM) to power the next generation of AI applications and agents. Palmyra X 004 offers the most accurate and reliable way for enterprises to enable actions within generative AI applications. The new frontier model ranks at the top of Berkeley’s Tool Calling Leaderboard by a significant margin, besting model providers including OpenAI, Anthropic, Meta, and Google. It also debuted as one of the top-ranked models on Stanford University’s Holistic Evaluation of Language Models (HELM) benchmark.
Writer successfully harnessed the power of synthetic data to train Palmyra X 004 at a fraction of the cost reported by major AI labs, while achieving groundbreaking accuracy and performance. To precision train its models, Writer curates and manufactures structured data via a proprietary LLM, Instruct-Adapt-X, and leverages an early stopping mechanism its team developed to achieve proficiency with a small percentage of the data used to train other frontier models.
"Unlocking the full potential of generative AI for our powerhouse customers like Intuit, Uber, L’Oreal, and Accenture requires the ability to execute complex actions and workflows,” said Writer CEO and co-founder May Habib. “Writer is pioneering a new era of LLM advancement that’s being overlooked by big tech, as they leverage their resources for sheer training data volume. Larger datasets are hitting their ceiling; the future belongs to precision training and architectural innovation—areas that we have long prioritized. Expect Writer’s approach to keep outperforming the market while meeting critical enterprise requirements.”
Palmyra X 004: Empowering Enterprises with Actionable AI
Writer's unique approach to model training has enabled it to produce models with state-of-the-art reasoning capabilities. Palmyra X 004 comes with a suite of new skills, such as:
- Taking action in systems external to the LLM via tool calling
- Automatic data integration with built-in retrieval augmented generation (RAG), including chain-of-thought and source transparency
- Code generation and deployment
- A 128K context window
As enterprises accelerate adoption of generative AI, there is increasing demand for AI-enabled applications that go beyond data analysis and text generation. With Palmyra X 004’s new tool calling capabilities, AI assistants and agents built with Writer can now be fully customized to interact with external systems, including performing tasks, fetching and analyzing data, deploying code, completing transactions, and executing workflows.
Palmyra Family of Models Bests the Biggest AI Labs
Palmyra X 004 is now the most accurate and affordable model for tool calling and API selection, outperforming competitors by nearly 20% on Berkeley’s Tool Calling Leaderboard. This benchmark tests the performance of LLMs using tool calling in real-world scenarios. It measures a model’s ability to select the correct tool to use, determine which API to call, and successfully execute a task based on a natural language input. Palmyra X 004’s overall score of 78.76% shows its capacity to reason through multiple actions quickly, accurately, and reliably.
The model also achieved impressive marks on the latest release of Stanford University’s Holistic Evaluation of Language Models (HELM). Palmyra X 004 debuted as one of the world’s top 10 models on both HELM Lite, a holistic framework for evaluating foundation models, and HELM MMLU, which tests understanding across 57 subjects, scoring 86.1% and 81.3% respectively.
Writer has a four-year track record of model innovation, including open-source, closed environment, vision, and domain-specific models for industry verticals. Palmyra X 004 joins Writer’s widely popular Palmyra family of enterprise-grade LLMs, which support multilingual capabilities in 30+ languages and multi-modal inputs across images, audio, and video. Combined with the Writer full-stack generative AI platform — which includes integrated graph-based RAG technology, AI guardrails, and a suite of developer tools — Palmyra LLMs have enabled hundreds of enterprises to reinvent their workflows with generative AI.
Palmyra X 004 is available today to enterprises via its API, the Writer Framework, no-code apps, and its out-of-the-box Ask Writer chat experience. You can also utilize the power of 004 in Writer’s recently launched Slack integration.
About Writer
Writer is the full-stack generative AI platform delivering transformative ROI for the world’s leading enterprises. Its all-in-one solution makes it easy to deploy customized AI apps and workflows that accelerate growth, increase productivity, and enhance compliance. Designed to provide enterprise-grade accuracy, security, and efficiency, Writer’s suite of development tools is supported by Palmyra – Writer’s state-of-the-art family of LLMs – alongside its industry-leading graph-based RAG and customizable AI guardrails. Named one of the top 50 companies in AI by Forbes, Writer empowers hundreds of customers like Accenture, Intuit, L’Oreal, and Vanguard to transform the way they work. Founded in 2020 with offices in San Francisco, New York City, and London, Writer is backed by strategic investors, including ICONIQ Growth, Insight Partners, WndrCo, Balderton Capital, and Aspect Ventures. Learn more at writer.com.
