Dust

Dust is for designing and deploying large language model apps. With Dust, users can build powerful workflows on top of LLMs and semantic search, making it easier than ever to create and deploy complex language models. Dust allows users to chain LLM apps arbitrarily between calls to models, code execution, and queries to external services. One of the most impressive features of Dust is its ability to avoid overfitting by iterating on LLM app design on several inputs simultaneously. Users can design against models served by OpenAI, Cohere, AI21, and more, and switch models seamlessly. Dust also offers easy access to iterations, model outputs, and few-shot examples saved automatically, making it simple to keep track of version history. Dust also offers caching to speed up iterations and reduce costs with cached model interactions. Users can deploy to an API endpoint or use Dust directly, and fully managed semantic search engines can be queried from workflows. Dust also offers connections to Notion, Google Docs, and Slack, which are kept up-to-date automatically. Overall, Dust is an essential software for anyone looking to design and deploy large language model apps. Try it now and experience the power of Dust for yourself.