fix: exclude empty stop parameter from LLM completion params

This fixes issue #4149 where models like gpt-5.1 don't support the
stop parameter at all. Previously, an empty list was always passed
to the API, causing an error.

Changes:
- Only include stop in params when it has values (non-empty)
- Respect additional_drop_params to ensure retry logic works correctly
- Add tests to verify the fix

Co-Authored-By: João <joao@crewai.com>

docs/en/enterprise/guides/gmail-trigger.mdx

@@ -62,13 +62,13 @@ Test your Gmail trigger integration locally using the CrewAI CLI:
 crewai triggers list
 
 # Simulate a Gmail trigger with realistic payload
-crewai triggers run gmail/new_email
+crewai triggers run gmail/new_email_received
 ```
 
 The `crewai triggers run` command will execute your crew with a complete Gmail payload, allowing you to test your parsing logic before deployment.
 
 <Warning>
-  Use `crewai triggers run gmail/new_email` (not `crewai run`) to simulate trigger execution during development. After deployment, your crew will automatically receive the trigger payload.
+  Use `crewai triggers run gmail/new_email_received` (not `crewai run`) to simulate trigger execution during development. After deployment, your crew will automatically receive the trigger payload.
 </Warning>
 
 ## Monitoring Executions
@@ -83,6 +83,6 @@ Track history and performance of triggered runs:
 
 - Ensure Gmail is connected in Tools & Integrations
 - Verify the Gmail Trigger is enabled on the Triggers tab
-- Test locally with `crewai triggers run gmail/new_email` to see the exact payload structure
+- Test locally with `crewai triggers run gmail/new_email_received` to see the exact payload structure
 - Check the execution logs and confirm the payload is passed as `crewai_trigger_payload`
 - Remember: use `crewai triggers run` (not `crewai run`) to simulate trigger execution

docs/ko/enterprise/guides/gmail-trigger.mdx

@@ -62,13 +62,13 @@ CrewAI CLI를 사용하여 Gmail 트리거 통합을 로컬에서 테스트하
 crewai triggers list
 
 # 실제 payload로 Gmail 트리거 시뮬레이션
-crewai triggers run gmail/new_email
+crewai triggers run gmail/new_email_received
 ```
 
 `crewai triggers run` 명령은 완전한 Gmail payload로 크루를 실행하여 배포 전에 파싱 로직을 테스트할 수 있게 해줍니다.
 
 <Warning>
-  개발 중에는 `crewai triggers run gmail/new_email`을 사용하세요 (`crewai run`이 아님). 배포 후에는 크루가 자동으로 트리거 payload를 받습니다.
+  개발 중에는 `crewai triggers run gmail/new_email_received`을 사용하세요 (`crewai run`이 아님). 배포 후에는 크루가 자동으로 트리거 payload를 받습니다.
 </Warning>
 
 ## Monitoring Executions
@@ -83,6 +83,6 @@ Track history and performance of triggered runs:
 
 - Ensure Gmail is connected in Tools & Integrations
 - Verify the Gmail Trigger is enabled on the Triggers tab
-- `crewai triggers run gmail/new_email`로 로컬 테스트하여 정확한 payload 구조를 확인하세요
+- `crewai triggers run gmail/new_email_received`로 로컬 테스트하여 정확한 payload 구조를 확인하세요
 - Check the execution logs and confirm the payload is passed as `crewai_trigger_payload`
 - 주의: 트리거 실행을 시뮬레이션하려면 `crewai triggers run`을 사용하세요 (`crewai run`이 아님)

docs/pt-BR/enterprise/guides/gmail-trigger.mdx

@@ -62,13 +62,13 @@ Teste sua integração de trigger do Gmail localmente usando a CLI da CrewAI:
 crewai triggers list
 
 # Simule um trigger do Gmail com payload realista
-crewai triggers run gmail/new_email
+crewai triggers run gmail/new_email_received
 ```
 
 O comando `crewai triggers run` executará sua crew com um payload completo do Gmail, permitindo que você teste sua lógica de parsing antes do deployment.
 
 <Warning>
-  Use `crewai triggers run gmail/new_email` (não `crewai run`) para simular execução de trigger durante o desenvolvimento. Após o deployment, sua crew receberá automaticamente o payload do trigger.
+  Use `crewai triggers run gmail/new_email_received` (não `crewai run`) para simular execução de trigger durante o desenvolvimento. Após o deployment, sua crew receberá automaticamente o payload do trigger.
 </Warning>
 
 ## Monitoring Executions
@@ -83,6 +83,6 @@ Track history and performance of triggered runs:
 
 - Ensure Gmail is connected in Tools & Integrations
 - Verify the Gmail Trigger is enabled on the Triggers tab
-- Teste localmente com `crewai triggers run gmail/new_email` para ver a estrutura exata do payload
+- Teste localmente com `crewai triggers run gmail/new_email_received` para ver a estrutura exata do payload
 - Check the execution logs and confirm the payload is passed as `crewai_trigger_payload`
 - Lembre-se: use `crewai triggers run` (não `crewai run`) para simular execução de trigger

lib/crewai/src/crewai/cli/authentication/main.py

@@ -149,7 +149,9 @@ class AuthenticationCommand:
                 return
 
             if token_data["error"] not in ("authorization_pending", "slow_down"):
-                raise requests.HTTPError(token_data["error_description"])
+                raise requests.HTTPError(
+                    token_data.get("error_description") or token_data.get("error")
+                )
 
             time.sleep(device_code_data["interval"])
             attempts += 1

lib/crewai/src/crewai/llm.py

@@ -676,14 +676,13 @@ class LLM(BaseLLM):
         formatted_messages = self._format_messages_for_provider(messages)
 
         # --- 2) Prepare the parameters for the completion call
-        params = {
+        params: dict[str, Any] = {
             "model": self.model,
             "messages": formatted_messages,
             "timeout": self.timeout,
             "temperature": self.temperature,
             "top_p": self.top_p,
             "n": self.n,
-            "stop": self.stop,
             "max_tokens": self.max_tokens or self.max_completion_tokens,
             "presence_penalty": self.presence_penalty,
             "frequency_penalty": self.frequency_penalty,
@@ -702,6 +701,12 @@ class LLM(BaseLLM):
             **self.additional_params,
         }
 
+        # Only include stop if it has values and is not in additional_drop_params
+        # Some models (e.g., gpt-5.1) don't support the stop parameter at all
+        drop_params = self.additional_params.get("additional_drop_params", [])
+        if self.stop and "stop" not in drop_params:
+            params["stop"] = self.stop
+
         # Remove None values from params
         return {k: v for k, v in params.items() if v is not None}
 

lib/crewai/tests/test_llm.py

@@ -877,3 +877,62 @@ def test_validate_model_in_constants():
         LLM._validate_model_in_constants("anthropic.claude-future-v1:0", "bedrock")
         is True
     )
+
+
+def test_prepare_completion_params_excludes_empty_stop():
+    """Test that _prepare_completion_params excludes stop when it's empty.
+
+    This is a regression test for issue #4149 where models like gpt-5.1
+    don't support the stop parameter at all, and passing an empty list
+    would cause an error.
+    """
+    llm = LLM(model="gpt-4o", is_litellm=True)
+    # By default, stop is initialized to an empty list
+    assert llm.stop == []
+
+    params = llm._prepare_completion_params("Hello")
+    # stop should not be in params when it's empty
+    assert "stop" not in params
+
+
+def test_prepare_completion_params_includes_stop_when_provided():
+    """Test that _prepare_completion_params includes stop when it has values."""
+    llm = LLM(model="gpt-4o", stop=["Observation:"], is_litellm=True)
+    assert llm.stop == ["Observation:"]
+
+    params = llm._prepare_completion_params("Hello")
+    # stop should be in params when it has values
+    assert "stop" in params
+    assert params["stop"] == ["Observation:"]
+
+
+def test_prepare_completion_params_excludes_stop_when_in_drop_params():
+    """Test that _prepare_completion_params excludes stop when it's in additional_drop_params.
+
+    This ensures the retry logic works correctly when a model doesn't support stop.
+    """
+    llm = LLM(
+        model="gpt-4o",
+        stop=["Observation:"],
+        additional_drop_params=["stop"],
+        is_litellm=True,
+    )
+    assert llm.stop == ["Observation:"]
+
+    params = llm._prepare_completion_params("Hello")
+    # stop should not be in params when it's in additional_drop_params
+    assert "stop" not in params
+
+
+def test_prepare_completion_params_excludes_stop_with_existing_drop_params():
+    """Test that stop is excluded when additional_drop_params already has other params."""
+    llm = LLM(
+        model="gpt-4o",
+        stop=["Observation:"],
+        additional_drop_params=["another_param", "stop"],
+        is_litellm=True,
+    )
+
+    params = llm._prepare_completion_params("Hello")
+    # stop should not be in params
+    assert "stop" not in params