LLM Observability

This release addresses several critical bug fixes across various integrations and internal components, including memory leak resolutions and stability improvements for Python versions 3.9-3.12.3. It also resolves issues with span export, trace data handling, and LLM observability.

This release addresses several memory growth issues related to profiling and symbol database handling. It also resolves a span generation problem with OpenRouter requests when the OpenAI integration is active.

This release introduces several new features for LLM Observability, including improved tracing, experiment evaluation, and Git metadata integration. It also enhances API security features for Flask, FastAPI, and Starlette, and improves database and Kafka monitoring capabilities.

This release focuses on stability and data integrity, fixing several bugs across LLM Observability, runtime metrics, IAST, profiling, and tracing components.

This release focuses on stability by fixing a critical tracing dependency issue related to wrapt and cleaning up unnecessary logs in LLM Observability when running agentless.

This release introduces significant performance improvements to profiling via Cython compilation and adds extensive new features across MLFlow, AI Guard, Azure Durable Functions, and deep enhancements to LLM Observability, including Pydantic evaluation support and incremental experiment reporting. Process tags are now propagated across many components by default.

This release focuses on bug fixes across CI Visibility, LLM Observability, and Error Tracking, including performance improvements and stability enhancements.

This release fixes bugs related to OpenAI error handling and LLM Observability tag parsing, and adds URL origin override capability to LLM Observability.

This release introduces a new processor capability for LLM Observability to process span inputs and outputs, and fixes a bug related to UTF-8 character counting during span attribute truncation.