Describe your metrics in natural language and Kibana's AI chat generates ES|QL-backed dashboards and Vega-Lite charts, from scatter plots to conditional formatting and custom tooltips. Kibana's AI chat now builds full Elasticsearch Query Language–backed (ES|QL-backed) dashboards from a natural-language prompt in under a minute. 5, this moves to general availability (GA) ( technical preview in 9. 4 ) with error recovery that retries failed queries, 5x lower ES|QL generation costs through tiered model routing, and interactive filter controls.
This release also adds Vega-Lite chart creation through natural language, including scatter plots, box plots, conditional formatting, and custom tooltips that you'd normally have to hand-code in JSON. Automatic retry up to three times with query inspection and adjustment Natural-language creation, including scatter plots, box plots, conditional formatting, custom tooltips In the technical preview, the agent didn’t retry failed ES|QL queries. 5, it detects query errors and retries up to three times, inspecting each error and adjusting the query before giving up.
In practice, this eliminates the majority of empty-panel issues and produces dashboards that render correctly on the first try. Not every step in dashboard generation needs the same level of reasoning. 5, ES|QL generation routes through a lighter model by default and falls back to the primary model only when needed. If your connector uses Anthropic's Claude Opus 4. 8, that means 5x cheaper for ES|QL generation across all panels. Dashboards are only useful when they show the right window of data.
The agent now applies improved logic to pick a time range that makes sense for the data it's querying, unless the user asks for a specific time range. It considers the data's time distribution and adjusts accordingly, whether that means the last hour for a live incident or the last 90 days for a trend analysis, rather than defaulting to a fixed window. Dashboard creation now supports controls; that is, interactive filters that let viewers narrow a dashboard by field values without editing the underlying queries.