minor fixes

async_openai/external_client.py

@@ -51,6 +51,7 @@ def __init__(
 
         self.is_proxied = is_proxied if is_proxied is not None else \
            (self.provider.config.has_proxy and  '_noproxy' not in self.name)
+        # logger.info(f"External Provider Configured: {self.name} [Proxied: {self.is_proxied}]")
 
         self.settings: Optional[OpenAISettings] = kwargs.pop('settings', get_settings())
         self.client_callbacks: List[Callable] = []
@@ -151,6 +152,7 @@ def configure_client(self, **kwargs):
             },
             **extra_kwargs,
         )
+        # logger.info(f"External Configured: {self._client.base_url} [{self.name}]")
 
     def configure_routes(self, **kwargs):
         """

async_openai/manager.py

@@ -747,6 +747,7 @@ def select_external_client_names(
         else: client_names = self.external_client_names
         if noproxy_required: client_names = [c for c in client_names if 'noproxy' in c]
         else: client_names = [c for c in client_names if 'noproxy' not in c]
+        # logger.info(f'External Clients: {client_names}')
         if excluded_clients: client_names = [c for c in client_names if c not in excluded_clients]
         return list(set(client_names))
 
@@ -1302,6 +1303,8 @@ def create_embeddings(
         model: Optional[str] = None,
         auto_retry: Optional[bool] = True,
         strip_newlines: Optional[bool] = False,
+        headers: Optional[Dict[str, str]] = None,
+        noproxy_required: Optional[bool] = False,
         **kwargs,
     ) -> List[List[float]]:
         """
@@ -1322,20 +1325,20 @@ def create_embeddings(
             **kwargs
         )
         if strip_newlines: inputs = [i.replace('\n', ' ').strip() for i in inputs]
-        client = self.get_client(model = model, **kwargs)
+        client = self.get_client(model = model, noproxy_required = noproxy_required, **kwargs)
         if not client.is_azure:
-            response = client.embeddings.create(input = inputs, model = model, auto_retry = auto_retry, **kwargs)
+            response = client.embeddings.create(input = inputs, model = model, auto_retry = auto_retry, headers = headers, **kwargs)
             return response.embeddings
 
         embeddings = []
         # We need to split into batches of 5 for Azure
         # Azure has a limit of 5 inputs per request
         batches = split_into_batches(inputs, 5)
         for batch in batches:
-            response = client.embeddings.create(input = batch, model = model, auto_retry = auto_retry, **kwargs)
+            response = client.embeddings.create(input = batch, model = model, auto_retry = auto_retry, headers = headers, **kwargs)
             embeddings.extend(response.embeddings)
             # Shuffle the clients to load balance
-            client = self.get_client(model = model, azure_required = True, **kwargs)
+            client = self.get_client(model = model, azure_required = True, noproxy_required = noproxy_required, **kwargs)
         return embeddings
 
 
@@ -1345,6 +1348,8 @@ async def async_create_embeddings(
         model: Optional[str] = None,
         auto_retry: Optional[bool] = True,
         strip_newlines: Optional[bool] = False,
+        headers: Optional[Dict[str, str]] = None,
+        noproxy_required: Optional[bool] = False,
         **kwargs,
     ) -> List[List[float]]:
         """
@@ -1365,20 +1370,20 @@ async def async_create_embeddings(
             **kwargs
         )
         if strip_newlines: inputs = [i.replace('\n', ' ').strip() for i in inputs]
-        client = self.get_client(model = model, **kwargs)
+        client = self.get_client(model = model, noproxy_required = noproxy_required, **kwargs)
         if not client.is_azure:
-            response = await client.embeddings.async_create(input = inputs, model = model, auto_retry = auto_retry, **kwargs)
+            response = await client.embeddings.async_create(input = inputs, model = model, auto_retry = auto_retry, headers = headers, **kwargs)
             return response.embeddings
 
         embeddings = []
         # We need to split into batches of 5 for Azure
         # Azure has a limit of 5 inputs per request
         batches = split_into_batches(inputs, 5)
         for batch in batches:
-            response = await client.embeddings.async_create(input = batch, model = model, auto_retry = auto_retry, **kwargs)
+            response = await client.embeddings.async_create(input = batch, model = model, auto_retry = auto_retry, headers = headers, **kwargs)
             embeddings.extend(response.embeddings)
             # Shuffle the clients to load balance
-            client = self.get_client(model = model, azure_required = True, **kwargs)
+            client = self.get_client(model = model, azure_required = True, noproxy_required = noproxy_required, **kwargs)
         return embeddings
 
     acreate_embeddings = async_create_embeddings

async_openai/types/functions.py

@@ -410,14 +410,15 @@ async def arun_chat_function(
         functions = functions or self.functions
         try:
             if headers: chat.client.headers.update(headers)
+            function_call = "auto" if function_name and function_name == 'auto' else {'name': function_name or self.function_name or self.name}
             return await chat.async_create(
                 model = model,
                 messages = messages,
                 functions = functions,
                 headers = headers,
                 auto_retry = True,
                 auto_retry_limit = 2,
-                function_call = {'name': function_name or self.name},
+                function_call = function_call,
                 header_cache_keys = ['Helicone-Cache-Enabled'],
                 **kwargs,
             )
@@ -480,13 +481,14 @@ def run_chat_function(
                 headers.update(property_meta)
         if headers: chat.client.headers.update(headers)
         functions = functions or self.functions
+        function_call = "auto" if function_name and function_name == 'auto' else {'name': function_name or self.function_name or self.name}
         return chat.create(
             messages = messages,
             functions = functions,
             headers = headers,
             auto_retry = True,
             auto_retry_limit = self.retry_limit,
-            function_call = {'name': function_name or self.name},
+            function_call = function_call,
             header_cache_keys=['Helicone-Cache-Enabled'],
             **kwargs,
         )

async_openai/utils/embedding.py

@@ -27,6 +27,8 @@ def _initialize_distance_dict(*args, **kwargs) -> Dict[str, Callable[..., float]
     """
     return {
         "cosine": spatial.distance.cosine,
+        "euclidean": spatial.distance.euclidean,
+        "inner_product": lambda x, y: -np.dot(x, y),
         "L1": spatial.distance.cityblock,
         "L2": spatial.distance.euclidean,
         "Linf": spatial.distance.chebyshev,

async_openai/version.py

@@ -1 +1 @@
-VERSION = '0.0.51rc1'
+VERSION = '0.0.51rc2'

tests/external_provider.py

@@ -1,6 +1,7 @@
 import os
 
-os.environ['TOGETHER_API_KEY'] = 'test123'
+# os.environ['TOGETHER_API_KEY'] = 'test123'
+os.environ['TOGETHER_API_KEYS'] = '[test1253, test4565]'
 
 from async_openai.utils.external_config import ExternalProviderSettings