OpenInference

OpenInference belongs in the database because standards matter in AI engineering, especially as teams try to avoid lock-in across tracing and observability tools. The official GitHub project positions OpenInference as OpenTelemetry instrumentation for AI observability, while the earlier specification repository explains the semantic model for inference and trace data. That makes OpenInference different from a platform like Opik or a benchmarking tool like GuideLLM. It is a standardization and instrumentation layer meant to make AI traces more portable and consistent.

It is a strong entry because many teams now rely on it indirectly through observability tooling even if they do not think of it as a top-level product. For the database, it should be framed honestly as an open standard and instrumentation toolkit rather than a paid platform. The economics are therefore simple: it is open source and free, and the cost discussion happens in the downstream observability stack that consumes the traces.