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
test(agent): name the exception a rewritten scan raises, not just the count it reports
Aashishc12f7013cc45
diff --git a/tests/test_predict_long.py b/tests/test_predict_long.pyindex 2dfce408..8325d2c2 100644--- a/tests/test_predict_long.py+++ b/tests/test_predict_long.py@@ -309,6 +309,7 @@ check("narrow/and two states is what the forward saw", a._forward_calls, [4])a = make_real_agent()grown, exc = _attempt(lambda: a.predict_long(LONG, Q, on_predict_start=lambda ctx: ctx.states.append("invented")))+check("grow/scoring one invented state is not an error", _kind(exc), None)check("grow/the hook's added state is scored",((grown or {}).get("usage") or {}).get("windows", "<absent>"), nwin + 1)check("grow/the caller's split is the one that was extended", a._forward_calls, [2 * (nwin + 1)])