revert: remove adapters / intrinsics / alora / lora from openai code (#543)

* revert: remove adapters / intrinsics / alora / lora from openai code

openai backend no longer supports intrinsics for vllm; vllm branch to
support aloras was closed with prejudice.

* fix: fix generate_from_raw with format for newer vllm server versions

* fix: remove unnecessary vllm flags

* doc: removed more mentions to VLLM_USE_V1=0

* refactor: make openai vllm test 'normal'

* fix: vllm server complains numpy version

---------

Co-authored-by: Masataro Asai <guicho2.71828@gmail.com>

Author: jakelorocco

docs/examples/conftest.py

@@ -264,7 +264,6 @@ def _run_vllm_examples_isolated(session, vllm_files: list[str]) -> int:
 
     # Set environment variables for vLLM
     env = os.environ.copy()
-    env["VLLM_USE_V1"] = "0"
     env["PYTORCH_CUDA_ALLOC_CONF"] = "expandable_segments:True"
 
     all_passed = True

docs/examples/intrinsics/intrinsics.py

@@ -3,26 +3,12 @@
 import mellea.stdlib.functional as mfuncs
 from mellea.backends.adapters.adapter import AdapterType, GraniteCommonAdapter
 from mellea.backends.huggingface import LocalHFBackend
-from mellea.backends.openai import OpenAIBackend, _ServerType
 from mellea.stdlib.components import Intrinsic, Message
 from mellea.stdlib.context import ChatContext
 
 # This is an example for how you would directly use intrinsics. See `mellea/stdlib/intrinsics/rag.py`
 # for helper functions.
 
-# Create the backend. Example for a VLLM Server. Commented out in favor of the hugging face code for now.
-# # Assumes a locally running VLLM server.
-# backend = OpenAIBackend(
-#     model_id="ibm-granite/granite-4.0-micro",
-#     base_url="http://0.0.0.0:8000/v1",
-#     api_key="EMPTY",
-# )
-
-# # If using a remote VLLM server, utilize the `test/backends/test_openai_vllm/serve.sh`
-# # script with `export VLLM_DOWNLOAD_RAG_INTRINSICS=True`. This will download the granite_common
-# # adapters on the server.
-# backend._server_type = _ServerType.REMOTE_VLLM
-
 backend = LocalHFBackend(model_id="ibm-granite/granite-3.3-8b-instruct")
 
 # Create the Adapter. GraniteCommonAdapter's default to ALORAs.

mellea/backends/adapters/__init__.py

@@ -5,7 +5,6 @@
     AdapterType,
     GraniteCommonAdapter,
     LocalHFAdapter,
-    OpenAIAdapter,
     fetch_intrinsic_metadata,
     get_adapter_for_intrinsic,
 )
@@ -15,7 +14,6 @@
     "AdapterType",
     "GraniteCommonAdapter",
     "LocalHFAdapter",
-    "OpenAIAdapter",
     "fetch_intrinsic_metadata",
     "get_adapter_for_intrinsic",
 ]

mellea/backends/adapters/adapter.py

@@ -38,26 +38,6 @@ def __init__(self, name: str, adapter_type: AdapterType):
         """set when the adapter is added to a backend"""
 
 
-class OpenAIAdapter(Adapter):
-    """Adapter for OpenAIBackends."""
-
-    @abc.abstractmethod
-    def get_open_ai_path(
-        self,
-        base_model_name: str,
-        server_type: _ServerType = _ServerType.LOCALHOST,
-        remote_path: str | None = None,
-    ) -> str:
-        """Returns the path needed to load the adapter.
-
-        Args:
-            base_model_name: the base model; typically the last part of the huggingface model id like "granite-3.3-8b-instruct"
-            server_type: the server type (ie LOCALHOST / OPENAI); usually the backend has information on this
-            remote_path: optional; used only if the server_type is REMOTE_VLLM; base path at which to find the adapter
-        """
-        ...
-
-
 class LocalHFAdapter(Adapter):
     """Adapter for LocalHFBackends."""
 
@@ -71,7 +51,7 @@ def get_local_hf_path(self, base_model_name: str) -> str:
         ...
 
 
-class GraniteCommonAdapter(OpenAIAdapter, LocalHFAdapter):
+class GraniteCommonAdapter(LocalHFAdapter):
     """Adapter for intrinsics that utilize the ``granite-common`` library."""
 
     def __init__(
@@ -149,35 +129,6 @@ def __init__(
         assert config_dict is not None  # Code above should initialize this variable
         self.config: dict = config_dict
 
-    def get_open_ai_path(
-        self,
-        base_model_name: str,
-        server_type: _ServerType = _ServerType.LOCALHOST,
-        remote_path: str | None = None,
-    ) -> str:
-        """Returns the path needed to load the adapter.
-
-        Args:
-            base_model_name: the base model; typically the last part of the huggingface
-                model id like "granite-3.3-8b-instruct"
-            server_type: the server type (ie LOCALHOST / OPENAI); usually the backend
-                has information on this
-            remote_path: optional; used only if the server_type is REMOTE_VLLM; base
-                path at which to find the adapter
-        """
-        if server_type == _ServerType.LOCALHOST:
-            path = self.download_and_get_path(base_model_name)
-        elif server_type == _ServerType.REMOTE_VLLM:
-            if remote_path is None:
-                remote_path = "rag-intrinsics-lib"
-            path = self.get_path_on_remote(base_model_name, remote_path)
-        else:
-            raise ValueError(
-                f"{self} not supported for OpenAIBackend with server_type: {server_type}"
-            )
-
-        return path
-
     def get_local_hf_path(self, base_model_name: str) -> str:
         """Returns the path needed to load the adapter.
 
@@ -207,11 +158,6 @@ def download_and_get_path(self, base_model_name: str) -> str:
             )
         )
 
-    def get_path_on_remote(self, base_model_name: str, base_path: str) -> str:
-        """Assumes the files have already been downloaded on the remote server."""
-        # TODO: This will break when we switch to the new repo!!!
-        return f"./{base_path}/{self.name}/{self.adapter_type.value}/{base_model_name}"
-
 
 T = TypeVar("T")