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: give the ONNX truncation test a stub tokenizer
Nandakishor3c68ca2ccf6a
diff --git a/tests/test_audit_regressions.py b/tests/test_audit_regressions.pyindex 3c90c0b3..44d79ecd 100644--- a/tests/test_audit_regressions.py+++ b/tests/test_audit_regressions.py@@ -93,7 +93,15 @@ def test_onnx_long_list_state_keeps_newest_turn(monkeypatch):agent.cfg = {"max_len": 32, "head_max_len": 16}agent.temperature = [1.0, 1.0, 1.0]agent.temperature_by_options = {}- agent.tok = object()+ class _StubTokenizer:+ # _infer tokenizes the shared state once before build_sequence (#343), so the stub+ # needs the two things that step reads; build_sequence itself is patched out above.+ mask_token = "<mask>"++ def __call__(self, text, add_special_tokens=False):+ return {"input_ids": []}++ agent.tok = _StubTokenizer()try:ONNXAgent._infer(agent, [{"text": "old"}, {"text": "new"}], {"q": {"type": "noul", "instructions": "?"}