Lovable adds Opus 5.5, matching Opus 5 quality in a third to half fewer steps

On September 22, 2026, Lovable added Opus 5.5 and reported the same build quality as Opus 5 with a third to half fewer steps, fewer tokens and better verification discipline.

Archive entry: this covers an event from September 22, 2026, and we published it on November 8, 2026. The details come from sources dated at the time and may have changed since.

On September 22, 2026, Lovable added Opus 5.5 and published benchmark results for it in a blog post on Opus 5.5 in Lovable. Lovable said Opus 5.5 got the same results as Opus 5 while finishing in a third to half fewer steps.

What the benchmarks showed

On verification discipline, Opus 5.5 scored 4% to 6% ahead of Opus 5. The two models scored level on 0-to-1 building and on iterative code fixing.
Steps per task fell 48%, 26% and 26% at low, medium and high effort for 0-to-1 building, and 47%, 34% and 34% for iterative code fixing.
Input tokens per task fell 21% to 46% on 0-to-1 building and 27% to 36% on iterative code fixing.
Output tokens fell 37% to 64% on iterative code fixing and verification discipline.
Lovable said Opus 5.5 gathered context in one pass, patched several files per round and checked only as much as a change needed.

What it means for people who build with AI

For Lovable users, fewer steps meant faster long edits at the same quality. Token use fell as well, and verification discipline scored 4% to 6% higher, so the speed gain did not cost quality in Lovable's measurements.

How Lovable measured it

Each task was run at least three times at equal reasoning effort, with the same code and the same judges. Lovable reported every difference as statistically significant at the 95% level.

Sources

Faster builds, same quality: Opus 5.5 now in Lovable (Lovable, September 22, 2026, official)

Written with AI assistance, strictly from the sources listed above. Before it went live, automated checks verified its dates, sources and every language edition.