LLMStack

LLMStack is an open-source, no-code platform designed to help users build generative AI applications, chatbots, workflows, and agents using their own data. The application takes the form of a platform that supports model chaining across major providers and allows teams to connect various external data sources. It is aimed at users looking to create customized AI workflows and collaborate on app development without writing code.

• Build generative AI applications, chatbots, and agents using your own data
• Model chaining supporting major providers like OpenAI, Cohere, Stability AI, and Hugging Face
• Import and connect data sources including Web URLs, Sitemaps, PDFs, Audio, PPTs, Google Drive, and Notion imports
• Collaborative app building with viewer and collaborator roles
• Granular permission model to share apps publicly or restrict access to specific individuals

Pros

• Supports a wide variety of native data sources such as Google Drive, Notion, PDFs, and web URLs for connecting external data to LLMs
• Provides flexibility in model selection through integration with major providers like OpenAI, Cohere, Stability AI, and Hugging Face
• Enables team collaboration with role-based permissions for building and sharing applications

Cons

• Pricing is not published on the page and requires contacting the provider or checking the cloud offering
• Requires managing or deploying the platform via its open-source codebase or cloud offering
• Relies on external model providers for underlying AI capabilities

LLMStack fits developers, teams, and productivity-focused users who want to construct custom AI agents, chatbots, and workflows using their own data sources without writing code. It is a poor fit for users looking for an entirely offline solution without external model connections, or those requiring pre-packaged pricing tiers directly on the website.

The pricing model is open-source with a cloud offering available, but specific pricing tiers and costs are not published on the page. Users must contact them for pricing.

Unlike generic chatbot builders, LLMStack combines an open-source, no-code approach with support for chaining multiple models from providers like OpenAI, Cohere, Stability AI, and Hugging Face. It integrates a wide variety of specific data sources such as Notion, Google Drive, and sitemaps directly into custom workflows, paired with a granular permission system for team collaboration.

OpenAI, Cohere, Stability AI, Hugging Face, Web URLs, Sitemaps, PDFs, Audio, PPTs, Google Drive, and Notion

LLMStack is worth trying for developers and teams looking for an open-source, no-code platform to build AI agents and chatbots with their own data. The inclusion of model chaining, collaborative roles, and diverse data source integrations makes it a versatile tool for workflow automation. However, because pricing details are not published and must be requested directly, buyers must contact the provider to evaluate total cost.