Highlights:

  • Writer creates a range of large language models designed for the enterprise market. Its most advanced model, Palmyra X 004.
  • Writer offers Palmyra X 004 along with domain-specific models, including Palmyra Fin, which is optimized for financial tasks like identifying investment risks and explaining market trends.

Recently, the artificial intelligence startup Writer Inc. secured USD 200 million in Series C funding, increasing its valuation to USD 1.9 billion.

The funding round was led by Premji Invest, Radical Ventures, and ICONIQ Growth, with participation from over half a dozen additional investors, including the venture capital arms of Salesforce Inc., Adobe Inc., IBM Corp., and Workday Inc. This latest raise follows a USD 100 million Series B round about a year ago, which reportedly valued Writer between USD 500 million and USD 750 million.

San Francisco-based Writer specializes in developing large language models tailored for the enterprise market. The company’s most advanced model, Palmyra X 004, was launched last month and significantly outperforms GPT-4o in tool use, making it more reliable for tasks that involve interacting with external applications like databases.

Palmyra X 004 can handle prompts containing up to 128,000 tokens, where a token represents a small unit of data, such as a few characters or numbers. This large language model can generate text, write code, and analyze user-provided datasets to identify valuable patterns.

Writer offers Palmyra X 004 along with a variety of domain-specific models. One of these, Palmyra Fin, is designed for financial tasks like identifying investment risks and explaining market trends. Another model, Palmyra Med, aims to streamline the daily tasks of healthcare professionals.

The company offers its large language models alongside a service called Knowledge Graph, a retrieval-augmented generation (RAG) tool. This service allows the models to integrate data from external sources into their prompt responses. With Knowledge Graph, AI can access information from PDFs, presentations, charts, and other files.

Behind the scenes, the service uses a large language model to determine the relationships between the records provided by the user. Writer claims that this method reduces the likelihood of AI models generating hallucinations. Additionally, the company states that Knowledge Graph enhances the models’ ability to answer complex questions more effectively.

“Our reasoning engine powers AI apps that have structured inputs and outputs, allowing you to fit Writer seamlessly into an existing workstream. It also powers agents that can work with other systems, think through complex tasks, and analyze your company data,” May Habib, Writer co-founder and Chief Executive Officer wrote in a blog post.

Writer offers application programming interfaces (APIs) that allow developers to integrate its LLMs and Knowledge Graph into their services. For business users, the company provides a no-code tool that enables the creation of AI applications without any programming. Users can design the application’s interface using prebuilt components and customize elements like the style of prompt responses.

Writer offers application programming interfaces (APIs) that allow developers to integrate its LLMs and Knowledge Graph into their services. For business users, the company provides a no-code tool that enables the creation of AI applications without any programming. Users can design the application’s interface using prebuilt components and customize elements like the style of prompt responses.

To enable customers to quickly utilize its LLMs, Writer offers over half a dozen prebuilt AI applications. Some are versatile tools designed for tasks like code generation, while others are more specialized, such as an app that analyzes marketing materials from financial institutions to ensure regulatory compliance.

Writer intends to use the funds from its latest funding round to enhance its feature set. As part of this expansion, the company plans to launch additional AI applications for the healthcare, retail, and financial sectors. Writer also aims to broaden its lineup of large language models, focusing on those better equipped for reasoning tasks.