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(evals): the policy refusal names which policy is invalid, not just that it is a usage error
Aashish792b548eb3b6
diff --git a/tests/test_evals.py b/tests/test_evals.pyindex dc536e1d..b35833bb 100644--- a/tests/test_evals.py+++ b/tests/test_evals.py@@ -3472,6 +3472,7 @@ def test_cli_rejects_invalid_policy_json_and_version(monkeypatch, tmp_path, capserr = capsys.readouterr().errassert "Traceback" not in err, "exit 2 with a traceback is the shape this refuses"assert err.startswith("laya-evals: "), "a usage error, named as one: %r" % err+ assert "gate policy" in err, "and named as *which* usage error: %r" % errassert not built