If you want a system to orchestrate AI agents and LLM flows with a high level of automation without having to write much code, Flowise is a top contender. It's an open-source low-code platform that lets developers build custom LLM orchestration flows and AI agents through a graphical interface. With more than 100 integrations, including Langchain and LlamaIndex, Flowise can handle a lot of work like generating product catalogs and querying SQL databases. It can be self-hosted on AWS, Azure and GCP, and has lots of documentation and an active community.
Another top contender is Dify, an open-source foundation for generative AI apps. Dify includes a visual Orchestration Studio for designing AI apps and tools for secure data pipelines, prompt design and model tuning. It includes customizable LLM agents and a way to quickly deploy AI assistants, and it can be used securely on-premise. Dify's flexibility and active development make it a good choice for businesses and individuals who want to use AI securely and efficiently.
Weave is another no-code AI workflow builder that lets you automate processes by combining multiple LLMs without any programming. It includes easy dataflow management, app-ready deployment and low-cost AI models. Weave can be used for tasks like generating dialogue, creating chatbots and automating written content, and it has tiered pricing with a free tier.
If you're looking for something more focused, AirOps is a powerful LLM workflow builder that combines user data with more than 40 AI models. It can handle large-scale workflows with error tracking and human oversight, and it can be deployed and scaled with Zapier and webhooks. AirOps is particularly good for e-commerce, marketplaces and content teams, and it has a range of pricing tiers for different business needs.