Non-autoregressive System 1 decision engine. Typed choice, score and yes/no decisions over any text in a single forward pass, in 100+ languages, with a router that picks the right checkpoint per request.
Python calibrationclassificationdecision-modelhuggingfacejevmodernbertmultilingualnlppythonpytorchroutingtyped-decisionszero-shot
docs: set LAYA_DEFAULT_MODEL=english for mostly English traffic, not the reverse
Bruce-Yii4ef7ba15d873
diff --git a/README.md b/README.mdindex cdfe1c30..4bf4ddb8 100644--- a/README.md+++ b/README.md@@ -746,7 +746,7 @@ Configuration is by environment variable: `LAYA_HOST`, `LAYA_PORT`,`LAYA_THREADS` (cap torch intra-op threads for CPU inference — keep at or belowphysical cores), `LAYA_AUTO_TASK`, `LAYA_DEFAULT_MODEL` (the checkpoint a state with nolanguage evidence falls back to, `multilingual` by default; set it to `english` when most of-your traffic is not English), `LAYA_MAX_LOADED` (checkpoints resident at+your traffic is English), `LAYA_MAX_LOADED` (checkpoints resident atonce, 2 by default; raise it to 3 when `LAYA_AUTO_TASK` makes a third onereachable on demand, or the server rebuilds one every time routing switches),`LAYA_IDLE_UNLOAD_SECONDS` (unload idle checkpoints; `0` disables it, `300` frees device memory