Unstructured's Commercial SaaS API
Blog post from Unstructured
Since its launch in 2022, the Unstructured library has experienced significant growth, reaching over 4 million downloads and integration into nearly 10,000 public GitHub repositories and various LLM-powered products. To address challenges related to processing multiple document types and managing complex dependencies, the company is enhancing its commercial API offerings, providing three options: a commercial SaaS API, a capped free-tier SaaS API, and Marketplace APIs for Azure and AWS allowing users to operate the API within their own VPCs. The commercial SaaS API offers dedicated cloud infrastructure for improved data security and operates on a pay-as-you-go basis, while the free-tier API will now collect user-uploaded documents for model training unless users opt for the commercial API. New features, such as advanced chunking options and support for audio formats, will be exclusive to API offerings, while the open source library will be maintained primarily as a prototyping tool. The company encourages user engagement through social media and direct communication for feedback and collaboration.