WeaveA no-code AI workflow builder that enables businesses to automate operations by integrating multiple large language models (LLMs) and connecting prompts seam...

W&B Weave is an observability and evaluation platform that helps track and improve large language model (LLM) applications. Weave provides tools to trace, collect metrics, and evaluate application responses using LLM judges and custom scorers. Key features include tracing sessions, LLM calls, and tool calls, as well as manual instrumentation of custom agents. The platform supports integrations with popular SDKs and harnesses, as well as custom agent observability. Weave provides Python and TypeScript libraries for installing and using the platform. It is hosted on Weights & Biases (W&B), requiring a W&B account and API key for authentication. Users can trace calls to LLMs, review inputs and outputs, and view agent metrics in the Weave UI. While Weave facilitates automation and evaluation of LLM applications, it is not a no-code AI workflow builder as suggested by its name.