drop default api url for host of clients

.env.example

@@ -1,2 +1,2 @@
-TEST_API_URL=https://test.api.aleph-alpha.com
+TEST_API_URL=https://inference-api.your-domain.com
 TEST_TOKEN=

.github/workflows/integration.yml

@@ -26,5 +26,5 @@ jobs:
         run: |
           poetry run pytest
         env:
-          TEST_API_URL: https://api.aleph-alpha.com
+          TEST_API_URL: ${{ secrets.TEST_API_URL }}
           TEST_TOKEN: ${{ secrets.AA_API_TOKEN }}

Changelog.md

@@ -1,5 +1,10 @@
 # Changelog
 
+## 8.0.0
+
+- Remove default value for `host` parameter in `Client` and `AsyncClient`. Passing a value for
+  the `host` is now required.
+
 ## 7.6.0
 
 - Add `instructable_embed` to `Client` and `AsyncClient`

Contributing.md

@@ -16,15 +16,15 @@ Tests can be run using pytest. Make sure to create a `.env` file with the follow
 
 ```env
 # test settings
-TEST_API_URL=https://test.api.aleph-alpha.com
+TEST_API_URL=https://inference-api.your-domain.com
 TEST_TOKEN=your_token
 ```
 
 Instead of a token username and password can be used.
 
 ```env
 # test settings
-TEST_API_URL=https://api.aleph-alpha.com
+TEST_API_URL=https://inference-api.your-domain.com
 TEST_USERNAME=your_username
 TEST_PASSWORD=your_password
 ```

README.md

@@ -21,9 +21,10 @@ from aleph_alpha_client import Client, CompletionRequest, Prompt
 client = Client(token=os.getenv("AA_TOKEN"))
 request = CompletionRequest(
     prompt=Prompt.from_text("Provide a short description of AI:"),
+    host="https://inference-api.your-domain.com",
     maximum_tokens=64,
 )
-response = client.complete(request, model="luminous-extended")
+response = client.complete(request, model="pharia-1-llm-7b-control")
 
 print(response.completions[0].completion)
 ```
@@ -38,9 +39,10 @@ from aleph_alpha_client import AsyncClient, CompletionRequest, Prompt
 async with AsyncClient(token=os.environ["AA_TOKEN"]) as client:
     request = CompletionRequest(
         prompt=Prompt.from_text("Provide a short description of AI:"),
+        host="https://inference-api.your-domain.com",
         maximum_tokens=64,
     )
-    response = client.complete_with_streaming(request, model="luminous-base")
+    response = client.complete_with_streaming(request, model="pharia-1-llm-7b-control")
 
     async for stream_item in response:
         print(stream_item)

aleph_alpha_client/aleph_alpha_client.py

@@ -129,7 +129,7 @@ class Client:
         token (string, required):
             The API token that will be used for authentication.
 
-        host (string, required, default "https://api.aleph-alpha.com"):
+        host (string, required):
             The hostname of the API host.
 
         hosting(string, optional, default None):
@@ -164,16 +164,16 @@ class Client:
 
     Example usage:
         >>> request = CompletionRequest(
-                prompt=Prompt.from_text(f"Request"), maximum_tokens=64
+                prompt=Prompt.from_text(f"Request"), host="https://inference-api.your-domain.com", maximum_tokens=64
             )
         >>> client = Client(token=os.environ["AA_TOKEN"])
-        >>> response: CompletionResponse = client.complete(request, "luminous-base")
+        >>> response: CompletionResponse = client.complete(request, "pharia-1-llm-7b-control")
     """
 
     def __init__(
         self,
         token: str,
-        host: str = "https://api.aleph-alpha.com",
+        host: str,
         hosting: Optional[str] = None,
         request_timeout_seconds: int = DEFAULT_REQUEST_TIMEOUT,
         total_retries: int = 8,
@@ -706,7 +706,7 @@ class AsyncClient:
         token (string, required):
             The API token that will be used for authentication.
 
-        host (string, required, default "https://api.aleph-alpha.com"):
+        host (string, required):
             The hostname of the API host.
 
         hosting(string, optional, default None):
@@ -740,15 +740,17 @@ class AsyncClient:
             Internal feature.
 
     Example usage:
-        >>> request = CompletionRequest(prompt=Prompt.from_text(f"Request"), maximum_tokens=64)
+        >>> request = CompletionRequest(
+                prompt=Prompt.from_text(f"Request"), host="https://inference-api.your-domain.com", maximum_tokens=64
+            )
         >>> async with AsyncClient(token=os.environ["AA_TOKEN"]) as client:
-                response: CompletionResponse = await client.complete(request, "luminous-base")
+                response: CompletionResponse = await client.complete(request, "pharia-1-llm-7b-control")
     """
 
     def __init__(
         self,
         token: str,
-        host: str = "https://api.aleph-alpha.com",
+        host: str,
         hosting: Optional[str] = None,
         request_timeout_seconds: int = DEFAULT_REQUEST_TIMEOUT,
         total_retries: int = 8,
@@ -846,7 +848,6 @@ async def _post_request(
         json_body = self._build_json_body(request, model)
 
         query_params = self._build_query_parameters()
-
         async with self.session.post(
             self.host + endpoint, json=json_body, params=query_params
         ) as response:

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "aleph-alpha-client"
-version = "7.6.0"
+version = "8.0.0"
 description = "python client to interact with Aleph Alpha api endpoints"
 authors = ["Aleph Alpha <[email protected]>"]
 license = "MIT"

tests/common.py

@@ -34,7 +34,7 @@ def model_name() -> str:
 
 @pytest.fixture(scope="session")
 def chat_model_name() -> str:
-    return "llama-3.1-70b-instruct"
+    return "llama-3.1-8b-instruct"
 
 
 @pytest.fixture(scope="session")

tests/test_embed.py

@@ -115,7 +115,7 @@ async def test_can_instructable_embed_with_async_client(
     )
 
     response = await async_client.instructable_embed(
-        request, model="Pharia-1-Embedding-4608-control"
+        request, model="pharia-1-embedding-4608-control"
     )
     assert response.model_version is not None
     assert response.embedding
@@ -239,7 +239,7 @@ def test_embed_instructable(sync_client: Client):
     )
 
     result = sync_client.instructable_embed(
-        request=request, model="Pharia-1-Embedding-4608-control"
+        request=request, model="pharia-1-embedding-4608-control"
     )
 
     assert result.model_version is not None

tests/test_qa.py

@@ -18,7 +18,6 @@ async def test_can_qa_with_async_client(async_client: AsyncClient):
         query="Who likes pizza?",
         documents=[Document.from_text("Andreas likes pizza.")],
     )
-
     response = await async_client.qa(request)
     assert len(response.answers) == 1
     assert response.answers[0].score > 0.0