Writer's Palmyra models have outperformed those by OpenAI, Cohere, Anthropic, Google, and Microsoft in three important tests using the Stanford HELM benchmarking system for AI models. In addition to the models, Writer also offers a "Knowledge Graph" and a layer of applications including chatbots and prebuilt templates. The company emphasizes its commitment to security and data privacy, with CEO May Habib stating that Writer's LLMs keep customer data private and never use it for model training. The company has reported a 10x revenue growth in the last two years and a net revenue retention rate over 150%.
Key takeaways:
- San Francisco startup Writer has raised $100 million in Series B funding to continue developing its secure generative AI platform for enterprises.
- Writer's proprietary AI models, Palmyra Small, Palmyra Base, and Palmyra Large, are designed to assist enterprises with various functions, including content creation, email drafting, audio transcription, and summarization.
- The Palmyra models are trained on over 1 trillion tokens of formal and business writing containing no IP or copyrighted content, and have outperformed models by OpenAI, Cohere, Anthropic, Google, Microsoft, and more in three important tests.
- Writer's platform also includes a Knowledge Graph that connects with a company's proprietary data, a suite of applications including chatbots and prebuilt templates, and user-friendly UI options, making it a competitive choice for enterprises looking for AI solutions.