For AI agents: a documentation index is available at the root level at /llms.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
New
- Evaluator results: LLM graders now provide a written rationale alongside each score, making it easier to understand why an evaluation passed or failed.
Improved
- Model catalog: updated available and retired models, refreshed provider pricing, and expanded model capability details.
- Model catalog loading: models now load on demand for each organization, with retry handling and protection against stale responses.
- Log ingestion: very large log payloads are now captured reliably without stalling the ingestion pipeline.
- Search: improved full-text search performance with a new ClickHouse text index.
- Request logs: queries that exceed memory or time limits now return a clear, actionable response instead of a raw database error.
Fixed
- Workflows: fixed an amplification issue that could cause workflows to perform unnecessary repeated work.
