fix: OpenAI base_url default and reasoning effort model option. (generative-computing#271)

* Adding a fix to pass `reasoning_effort` in conditionally

* adding tests

* Fixes generative-computing#274

* Adds GPT 5.1 model identifier.

* Changes OpenAI Backend default model_id to GPT 5.1.

This default is changed because the default base_url is also changed.

* Fixes bug: GenSlots did not work with OpenAI platform.

The OpenAI response_format only accepts a limited set of schemas and
will error out with a 400 if you do not follow their guidelines.

One of these guidelines is that additionalProperties is set and is set
to False.

This commit monkey-patches the response_format provided to OpenAI
platform backends, and leaves other OpenAI-"compatible" backends with
the existing default behavior. This is a debatable choice.

See https://community.openai.com/t/api-rejects-valid-json-schema/906163 and https://platform.openai.com/docs/guides/structured-outputs?api-mode=chat

* Adds inline documentation for OpenAI model options monkey patching.

* removes debug print stmt.

* adding a comment about reasoning_effort in openai sdk

* removing all instances of hf_model_id in openai backend

* removing apply_chat_template and adding assertions for env variable

* adding some tests for param checking

* changing env variable handling logic.

* base_url check is now a warning

* fix: change warning message in openai.py

* marking test as qualitative cause it's causing timeouts in github actions(I think)

---------

Co-authored-by: Nathan Fulton <gitcommit@nfulton.org>
Co-authored-by: jakelorocco <59755218+jakelorocco@users.noreply.github.com>

Author: 3 people

mellea/backends/model_ids.py

@@ -17,6 +17,7 @@ class ModelIdentifier:
     ollama_name: str | None = None
     watsonx_name: str | None = None
     mlx_name: str | None = None
+    openai_name: str | None = None
 
     hf_tokenizer_name: str | None = None  # if None, is the same as hf_model_name
 
@@ -134,9 +135,9 @@ class ModelIdentifier:
 
 QWEN3_14B = ModelIdentifier(hf_model_name="Qwen/Qwen3-14B", ollama_name="qwen3:14b")
 
-######################
-#### OpenAI models ###
-######################
+###########################
+#### OpenAI open models ###
+###########################
 
 OPENAI_GPT_OSS_20B = ModelIdentifier(
     hf_model_name="openai/gpt-oss-20b", ollama_name="gpt-oss:20b"
@@ -145,6 +146,12 @@ class ModelIdentifier:
     hf_model_name="openai/gpt-oss-120b", ollama_name="gpt-oss:120b"
 )
 
+###########################
+#### OpenAI prop models ###
+###########################
+
+OPENAI_GPT_5_1 = ModelIdentifier(openai_name="gpt-5.1")
+
 #####################
 #### Misc models ####
 #####################

mellea/backends/openai.py

@@ -6,6 +6,7 @@
 import functools
 import inspect
 import json
+import os
 from collections.abc import Callable, Coroutine
 from copy import deepcopy
 from enum import Enum
@@ -72,7 +73,7 @@ class OpenAIBackend(FormatterBackend, AdapterMixin):
 
     def __init__(
         self,
-        model_id: str | ModelIdentifier = model_ids.IBM_GRANITE_4_MICRO_3B,
+        model_id: str | ModelIdentifier = model_ids.OPENAI_GPT_5_1,
         formatter: Formatter | None = None,
         base_url: str | None = None,
         model_options: dict | None = None,
@@ -142,26 +143,38 @@ def __init__(
 
         self.default_to_constraint_checking_alora = default_to_constraint_checking_alora
 
-        self._model_id = model_id
         match model_id:
             case str():
-                self._hf_model_id = model_id
+                self._model_id = model_id
             case ModelIdentifier():
-                assert model_id.hf_model_name is not None, (
-                    "model_id is None. This can also happen if the ModelIdentifier has no hf_model_id name set."
+                assert model_id.openai_name is not None, (
+                    "model_id is None. This can also happen if the ModelIdentifier has no `openai_name` name set."
                 )
-                self._hf_model_id = model_id.hf_model_name
+                self._model_id = model_id.openai_name
 
-        if base_url is None:
-            self._base_url = "http://localhost:11434/v1"  # ollama
-        else:
-            self._base_url = base_url
-        if api_key is None:
-            self._api_key = "ollama"
-        else:
-            self._api_key = api_key
+        # Use provided parameters or fall back to environment variables
+        self._api_key = api_key
+        self._base_url = base_url
 
-        self._server_type = _server_type(self._base_url)
+        # Validate that we have the required configuration
+        if self._api_key is None and os.getenv("OPENAI_API_KEY") is None:
+            raise ValueError(
+                "OPENAI_API_KEY or api_key is required but not set. Please either:\n"
+                "  1. Set the environment variable: export OPENAI_API_KEY='your-key-here'\n"
+                "  2. Pass it as a parameter: OpenAIBackend(api_key='your-key-here')"
+            )
+
+        if self._base_url is None and os.getenv("OPENAI_BASE_URL") is None:
+            FancyLogger.get_logger().warning(
+                "OPENAI_BASE_URL or base_url is not set.\n"
+                "The openai SDK is going to assume that the base_url is `https://api.openai.com/v1`"
+            )
+
+        self._server_type: _ServerType = (
+            _server_type(self._base_url)
+            if self._base_url is not None
+            else _ServerType.OPENAI
+        )  # type: ignore
 
         self._openai_client_kwargs = self.filter_openai_client_kwargs(**kwargs)
 
@@ -668,14 +681,38 @@ async def _generate_from_chat_context_standard(
 
         extra_params: dict[str, Any] = {}
         if _format is not None:
-            extra_params["response_format"] = {
-                "type": "json_schema",
-                "json_schema": {
-                    "name": _format.__name__,
-                    "schema": _format.model_json_schema(),
-                    "strict": True,
-                },
-            }
+            if self._server_type == _ServerType.OPENAI:
+                # The OpenAI platform requires that additionalProperties=False on all response_format schemas.
+                # However, not all schemas generates by Mellea include additionalProperties.
+                # GenerativeSlot, in particular, does not add this property.
+                # The easiest way to address this disparity between OpenAI and other inference providers is to
+                # monkey-patch the response format exactly when we are actually using the OpenAI server.
+                #
+                # This only addresses the additionalProperties=False constraint.
+                # Other constraints we should be checking/patching are described here:
+                # https://platform.openai.com/docs/guides/structured-outputs?api-mode=chat
+                monkey_patched_response_schema = _format.model_json_schema()
+                monkey_patched_response_schema["additionalProperties"] = False
+                extra_params["response_format"] = {
+                    "type": "json_schema",
+                    "json_schema": {
+                        "name": _format.__name__,
+                        "schema": monkey_patched_response_schema,
+                        "strict": True,
+                    },
+                }
+            else:
+                FancyLogger().get_logger().warning(
+                    "Mellea assumes you are NOT using the OpenAI platform, and that other model providers have less strict requirements on support JSON schemas passed into `format=`. If you encounter a server-side error following this message, then you found an exception to this assumption. Please open an issue at github.com/generative_computing/mellea with this stack trace and your inference engine / model provider."
+                )
+                extra_params["response_format"] = {
+                    "type": "json_schema",
+                    "json_schema": {
+                        "name": _format.__name__,
+                        "schema": _format.model_json_schema(),
+                        "strict": True,
+                    },
+                }
 
         # Append tool call information if applicable.
         tools: dict[str, Callable] = dict()
@@ -701,15 +738,21 @@ async def _generate_from_chat_context_standard(
         formatted_tools = convert_tools_to_json(tools)
         use_tools = len(formatted_tools) > 0
 
+        # Build optional reasoning parameters
+        # NOTE: the openai SDK doesn't like it if you pass `reasoning_effort` param to a non-reasoning model e.g. gpt4o
+        reasoning_params = {}
+        if thinking is not None:
+            reasoning_params["reasoning_effort"] = thinking
+
         chat_response: Coroutine[
             Any, Any, ChatCompletion | openai.AsyncStream[ChatCompletionChunk]
         ] = self._async_client.chat.completions.create(
-            model=self._hf_model_id,
+            model=self._model_id,
             messages=conversation,  # type: ignore
-            reasoning_effort=thinking,  # type: ignore
             tools=formatted_tools if use_tools else None,  # type: ignore
             # parallel_tool_calls=False, # We only support calling one tool per turn. But we do the choosing on our side so we leave this False.
             **extra_params,
+            **reasoning_params,  # type: ignore
             **self._make_backend_specific_and_remove(
                 model_opts, is_chat_context=ctx.is_chat_context
             ),
@@ -877,7 +920,7 @@ async def generate_from_raw(
         try:
             completion_response: Completion = (
                 await self._async_client.completions.create(
-                    model=self._hf_model_id,
+                    model=self._model_id,
                     prompt=prompts,
                     extra_body=extra_body,
                     **self._make_backend_specific_and_remove(
@@ -930,7 +973,10 @@ async def generate_from_raw(
     @property
     def base_model_name(self):
         """Returns the base_model_id of the model used by the backend. For example, `granite-3.3-8b-instruct` for `ibm-granite/granite-3.3-8b-instruct`."""
-        return self._hf_model_id.split("/")[1]
+        if "/" in self._model_id:
+            return self._model_id.split("/")[1]
+        else:
+            return self._model_id
 
     def add_adapter(self, adapter: OpenAIAdapter):
         """Adds the given adapter to the backend. Must not have been added to a different backend."""
@@ -1085,22 +1131,3 @@ def list_adapters(self) -> list[str]:
         :returns: list of adapter names that are currently registered with this backend
         """
         return list(self._loaded_adapters.keys())
-
-    def apply_chat_template(self, chat: list[dict[str, str]]):
-        """Apply the chat template for the model, if such a model is available (e.g., when it can deduce the huggingface model id)."""
-        from transformers import AutoTokenizer
-
-        if not hasattr(self, "_tokenizer"):
-            match _server_type(self._base_url):
-                case _ServerType.LOCALHOST:
-                    self._tokenizer: "PreTrainedTokenizer" = (  # noqa: UP037
-                        AutoTokenizer.from_pretrained(self._hf_model_id)
-                    )
-                case _ServerType.OPENAI:
-                    raise Exception(
-                        "apply_chat_template is called while targeting a server at openai.com. "
-                        "This is not supported --- openai.com does not support Activated Lora. "
-                        "Use a locally served vllm instance. "
-                    )
-
-        return self._tokenizer.apply_chat_template(chat, tokenize=False)

test/backends/test_openai_ollama.py

@@ -1,6 +1,7 @@
 # test/rits_backend_tests/test_openai_integration.py
 import asyncio
 import os
+from unittest.mock import patch
 
 import openai
 import pydantic
@@ -216,6 +217,80 @@ async def get_client_async():
     assert len(backend._client_cache.cache.values()) == 2
 
 
+async def test_reasoning_effort_conditional_passing(backend):
+    """Test that reasoning_effort is only passed to API when not None."""
+    from unittest.mock import AsyncMock, MagicMock, patch
+
+    ctx = ChatContext()
+    ctx = ctx.add(CBlock(value="Test"))
+
+    mock_response = MagicMock()
+    mock_response.choices = [MagicMock()]
+    mock_response.choices[0].message = MagicMock()
+    mock_response.choices[0].message.content = "Response"
+    mock_response.choices[0].message.role = "assistant"
+
+    # Test 1: reasoning_effort should NOT be passed when not specified
+    with patch.object(
+        backend._async_client.chat.completions, "create", new_callable=AsyncMock
+    ) as mock_create:
+        mock_create.return_value = mock_response
+        await backend.generate_from_chat_context(
+            CBlock(value="Hi"), ctx, model_options={}
+        )
+        call_kwargs = mock_create.call_args.kwargs
+        assert "reasoning_effort" not in call_kwargs, (
+            "reasoning_effort should not be passed when not specified"
+        )
+
+    # Test 2: reasoning_effort SHOULD be passed when specified
+    with patch.object(
+        backend._async_client.chat.completions, "create", new_callable=AsyncMock
+    ) as mock_create:
+        mock_create.return_value = mock_response
+        await backend.generate_from_chat_context(
+            CBlock(value="Hi"), ctx, model_options={ModelOption.THINKING: "medium"}
+        )
+        call_kwargs = mock_create.call_args.kwargs
+        assert call_kwargs.get("reasoning_effort") == "medium", (
+            "reasoning_effort should be passed with correct value when specified"
+        )
+
+
+def test_api_key_and_base_url_from_parameters():
+    """Test that API key and base URL can be set via parameters."""
+    backend = OpenAIBackend(
+        model_id="gpt-4", api_key="test-api-key", base_url="https://api.test.com/v1"
+    )
+    assert backend._api_key == "test-api-key"
+    assert backend._base_url == "https://api.test.com/v1"
+
+
+def test_parameter_overrides_env_variable():
+    """Test that explicit parameters override environment variables."""
+    with patch.dict(
+        os.environ,
+        {"OPENAI_API_KEY": "env-api-key", "OPENAI_BASE_URL": "https://api.env.com/v1"},
+    ):
+        backend = OpenAIBackend(
+            model_id="gpt-4",
+            api_key="param-api-key",
+            base_url="https://api.param.com/v1",
+        )
+        assert backend._api_key == "param-api-key"
+        assert backend._base_url == "https://api.param.com/v1"
+
+
+def test_missing_api_key_raises_error():
+    """Test that missing API key raises ValueError with helpful message."""
+    with patch.dict(os.environ, {}, clear=True):
+        with pytest.raises(ValueError) as exc_info:
+            OpenAIBackend(model_id="gpt-4", base_url="https://api.test.com/v1")
+        assert "OPENAI_API_KEY or api_key is required but not set" in str(
+            exc_info.value
+        )
+
+
 if __name__ == "__main__":
     import pytest
 

test/stdlib_intrinsics/test_rag/test_rag.py

@@ -184,6 +184,7 @@ def test_answer_relevance(backend):
     assert result == answer
 
 
+@pytest.mark.qualitative
 def test_answer_relevance_classifier(backend):
     """Verify that the first phase of the answer relevance flow behaves as expectee."""
     context, answer, docs = _read_input_json("answer_relevance.json")