LLMStack
A self-hosted no-code AI app builder that connects models to your data, supports Slack and Discord, and also offers hosted deployment through Promptly.
LLMStack is a self-hosted no-code builder for teams creating AI agents, chatbots and workflows around their own data and business processes. You can run it on your own infrastructure or use Promptly, its hosted cloud offering. Self-hosting runs the platform on your server; connected cloud model providers process requests outside it.
The builder lets you chain multiple models in one application and connect agents to internal or external tools. It supports OpenAI, Cohere, Stability AI and Hugging Face. Agents can search the web or browse websites, with uses such as research and sales outreach.
You can bring in PDFs, audio, presentations and other documents, along with web URLs, sitemaps, Google Drive and Notion content. LLMStack handles data preparation and stores it in an included vector database so applications can use the imported material. This makes it relevant to teams building assistants around company documents or other information they already maintain.
Apps can be public or restricted to selected people. Viewer and collaborator roles separate access from editing, while multiple organizations keep their users' data and AI workflows apart. Teams can build apps together and expose them through an HTTP API, or trigger workflows from Slack and Discord.
Windows deployment uses WSL2. Background jobs require Docker. Its source-available license restricts offering the software as a hosted or managed service without prior permission.
