v2.2
Improved
Inference
Semantic cache, second generation
Hit rate up from 31% to 54% on conversational workloads. Enabled per key.
Release
shipped
Version
v2.2
Released
September 4, 2026
Type
Improved
Product
Inference
Rolled out to every region and plan
What changed
Added
Intent matching for the semantic cache, so rephrased questions hit the same cached answer.
Changed
Hit rate on conversational workloads rose from 31% to 54%. Cache hits stay free and keep their trace.
Action needed
Enable the second-generation cache per API key in Settings, Inference. The first generation stays the default until Nov 30 2026.
Release notes
The semantic cache now matches on intent rather than wording. On conversational workloads the hit rate rose from 31% to 54%. Cache hits remain free.
Need a hand?
Questions about this release
The docs cover every flag and endpoint. For anything else, support answers within one business day.
Run your first agent in five minutes.
1M free tokens a month. No card, no sales call, production keys in two minutes.
$ npm install @lattis/sdk
$ export LATTIS_API_KEY=lt_••••••••
$ npx lattis run hello-agent
✓ plan · ✓ search_docs · ✓ answer
● run completed · 3 steps · 0.9 s