Some highlights of the platform:
- Chain multiple LLM models allowing for complex pipelines
- Includes a vector database and necessary connectors to help enrich LLM responses with private data
- App templates tailored to specific use cases to quickly build LLM apps in minutes
- Collaborative app editing and prompt engineering capabilities
- Streaming APIs, Slack and Discord integrations
- Multi-tenant ready for enterprise deployments with user management, org level keys etc.,
- Ability to use local open-source LLMs like Llama2 etc using LocalAI (https://localai.io)
Background: We started as a closed source prompt management platform early this year (https://trypromptly.com) and eventually landed as an Enterprise LLM apps platform. In the process, we learned how hard it is to sell a horizontal SaaS platform. That combined with the concerns around data privacy (both with us hosting data as well sending data to model providers like OpenAI), we found a lot of enterprises hesitant to signup for Promptly. We added the ability for the users to host their own vector databases which helped a bit. With the current pace of performance improvements in open source LLMs, enterprises now have the ability to run LLMs entirely locally or in their private clouds without suffering the loss in quality of output. This made us realize we can make Promptly self-hostable so enterprises can have a low-code apps platform on locally running LLMs without sending any data out. LLMStack came out of this realization.
Promptly is now powered by LLMStack which is entirely being developed in the open. Anyone can bring up LLMStack locally and have an experience similar to Promptly. We still have paid plans on Promptly for customers who don't want to deal with installations and offer on-prem installation support to enterprises that want to use LLMStack similar to Supabase, Airbyte etc. Our current focus with LLMStack is to make it work really well with open-source models. Happy to answer any questions.