v1.8.10

1.8.10 (2026-10-22)

Bug Fixes

• api-gateway: Report a queryRewrite that returns no query as a user error (#12342) (0e82b7c)
• cube-cli: Add per-run --output-path and --name-prefix to cube dbt (#12351) (6dc453b), closes cubedevinc/cubejs-enterprise#16867 cubedevinc/cubejs-enterprise#16867
• cube-cli: Send cube deploy --branch where the server reads it (#12350) (7fcee8a), closes cubedevinc/cubejs-enterprise#16868
• cubesql: Pull up subquery forms into one class per data source (#12354) (ec6c45a)
• cubesql: Push down variance and standard deviation aggregates (#12349) (0c695a5)
• cubestore: Accept XIRR convergence for large cash flows (#12235) (962a8c5)
• tesseract: Fix segment column recursion and ungrouped row pairing (#12222) (0ce1aa6)
• tesseract: Keep filters on plain members out of a multi-stage dimension select (#12227) (186fc62)
• tesseract: Report every missing grain dimension in one error (#12223) (1d75292)
• tesseract: Resolve multi-stage dimensions queried through a view (#12224) (5446b19)

Features

• docker: Security upgrade Node.js from 24.21.0 to 24.22.1 (#12353) (e037a7e)

Performance Improvements

• schema-compiler: Build rollup SQL evaluation query once per description (#12352) (5902215)
• tesseract: Reuse cached driver tools when extracting pre-aggregation scan ranges (#12339) (55b5383)