Engines

Generated page

This page is generated from the result rows and the override registry, and is not edited by hand. It is empty until the first 2026-11 rows land.

For every engine in every lane, this page will state what we run: the version, the container image digest where a row records one, the durability classes measured, the tiers, every setting a row carries, and each override of an engine's default with the reason. Everything will be read from the result rows and the override registry, so it cannot drift from what was measured. The rules these settings follow are on the Methodology page.

No row has been measured with the 2026-11 instrument yet, so there is nothing to list. The page is generated by benchmarks/experiments/gen_engines_doc.py in humemai/arcadedb-embedded-python from the clean rows of the 2026-11 instrument (the newest clean row of each cell, on the serial CPU set), together with the failed cells, which are listed as the result they are. It replaces this text when the first 2026-11 rows land. The earlier 2026-10 instrument was a development version; its rows were never published as a comparison and are not shown.