fix: remove unsupported include_stop_str_in_output NIM param (#95)

config/nim.py

@@ -23,7 +23,6 @@ class NimSettings(BaseModel):
 
     parallel_tool_calls: bool = True
     return_tokens_as_token_ids: bool = False
-    include_stop_str_in_output: bool = False
     ignore_eos: bool = False
 
     min_tokens: int = Field(0, ge=0)

providers/nvidia_nim/request.py

@@ -87,7 +87,6 @@ def build_request_body(request_data: Any, nim: NimSettings) -> dict:
     _set_extra(extra_body, "chat_template", nim.chat_template)
     _set_extra(extra_body, "request_id", nim.request_id)
     _set_extra(extra_body, "return_tokens_as_token_ids", nim.return_tokens_as_token_ids)
-    _set_extra(extra_body, "include_stop_str_in_output", nim.include_stop_str_in_output)
     _set_extra(extra_body, "ignore_eos", nim.ignore_eos)
     _set_extra(extra_body, "reasoning_effort", nim.reasoning_effort)
     _set_extra(extra_body, "include_reasoning", nim.include_reasoning)

tests/providers/test_nvidia_nim_request.py

@@ -2,6 +2,8 @@
 
 from unittest.mock import MagicMock
 
+import pytest
+
 from config.nim import NimSettings
 from providers.common.utils import set_if_not_none
 from providers.nvidia_nim.request import (
@@ -10,6 +12,23 @@ from providers.nvidia_nim.request import (
 )
 
 
+@pytest.fixture
+def req():
+    r = MagicMock()
+    r.model = "test"
+    r.messages = [MagicMock(role="user", content="hi")]
+    r.max_tokens = 100
+    r.system = None
+    r.temperature = None
+    r.top_p = None
+    r.stop_sequences = None
+    r.tools = None
+    r.tool_choice = None
+    r.extra_body = None
+    r.top_k = None
+    return r
+
+
 class TestSetIfNotNone:
     def test_value_not_none_sets(self):
         body = {}
@@ -45,57 +64,22 @@ class TestSetExtra:
 
 
 class TestBuildRequestBody:
-    def test_max_tokens_capped_by_nim(self):
-        """Request max_tokens exceeds nim.max_tokens -> capped."""
-        req = MagicMock()
-        req.model = "test"
-        req.messages = [MagicMock(role="user", content="hi")]
+    def test_max_tokens_capped_by_nim(self, req):
         req.max_tokens = 100000
-        req.system = None
-        req.temperature = None
-        req.top_p = None
-        req.stop_sequences = None
-        req.tools = None
-        req.tool_choice = None
-        req.extra_body = None
-        req.top_k = None
-
         nim = NimSettings(max_tokens=4096)
         body = build_request_body(req, nim)
         assert body["max_tokens"] == 4096
 
-    def test_presence_penalty_included_when_nonzero(self):
-        req = MagicMock()
-        req.model = "test"
-        req.messages = [MagicMock(role="user", content="hi")]
-        req.max_tokens = 100
-        req.system = None
-        req.temperature = None
-        req.top_p = None
-        req.stop_sequences = None
-        req.tools = None
-        req.tool_choice = None
-        req.extra_body = None
-        req.top_k = None
-
+    def test_presence_penalty_included_when_nonzero(self, req):
         nim = NimSettings(presence_penalty=0.5)
         body = build_request_body(req, nim)
         assert body["presence_penalty"] == 0.5
 
-    def test_parallel_tool_calls_included(self):
-        req = MagicMock()
-        req.model = "test"
-        req.messages = [MagicMock(role="user", content="hi")]
-        req.max_tokens = 100
-        req.system = None
-        req.temperature = None
-        req.top_p = None
-        req.stop_sequences = None
-        req.tools = None
-        req.tool_choice = None
-        req.extra_body = None
-        req.top_k = None
+    def test_include_stop_str_in_output_not_sent(self, req):
+        body = build_request_body(req, NimSettings())
+        assert "include_stop_str_in_output" not in body.get("extra_body", {})
 
+    def test_parallel_tool_calls_included(self, req):
         nim = NimSettings(parallel_tool_calls=False)
         body = build_request_body(req, nim)
         assert body["parallel_tool_calls"] is False