About Swirl Search
SWIRL is an enterprise search and answer platform that returns cited AI answers by searching a company's existing systems live, as the signed in user, without copying content into a central index. Version 5 is positioned at regulated organizations such as law firms, government bodies, pharma companies and universities.
Because nothing is copied, permissions stay enforced by each source system, deletion happens at the source, and no new store holds the data. A single query runs in parallel across connected systems, groups document versions, shows the approved version first and links each claim back to its source document. The platform lists more than 150 connectors, including Microsoft 365, SharePoint, Salesforce, ServiceNow, GitHub, Confluence, Jira, Google Drive, BigQuery and Snowflake, and it works with any LLM the customer brings. It exposes an MCP server and a REST API so agents such as Claude, Copilot or ChatGPT receive the same cited answers under the same permissions. Developers and OEMs can build on it through SWIRL for Embedding. The project began as an open source release under Apache 2.0 in 2023.
Pricing is an annual platform license. Departmental is $24K per year for one business unit, Business is $72K per year for multiple units, and Enterprise starts at $150K per year with custom scope. Multi year prepay is stated to save about 15 to 20 percent. The Semantic Cache add on is licensed separately. Evaluation is a free 30 minute guided stand up or a $12,000 paid pilot of 60 to 90 days, credited to the first annual contract. An OEM and source code license starts at $60K per year plus revenue share.