Update version 1.0.1

README.md

@@ -3,10 +3,10 @@ Copyright (c) 2022-present, FriendliAI Inc. All rights reserved.
 -->
 
 <p align="center">
-  <img src="https://docs.friendli.ai/img/logo.svg" width="80%" alt="system">
+  <img src="https://docs.friendli.ai/img/favicon.svg" width="30%" alt="Friendli Logo">
 </p>
 
-<h2><p align="center">Supercharge Generative AI Serving 🚀</p></h2>
+<h2><p align="center">Supercharge Generative AI Serving with Friendli 🚀</p></h2>
 
 <p align="center">
   <a href="https://github.com/friendliai/friendli-client/actions/workflows/ci.yaml">
@@ -57,12 +57,13 @@ Check out [Friendli Client Docs](https://docs.friendli.ai/) to learn more.
 pip install friendli-client
 ```
 
-If you have a Hugging Face checkpoint and want to convert it to a Friendli-compatible format with applying quantization, you need to install the package with the necessary machine learing library (`mllib`) dependencies.
-In this case, install the package with the following command:
-
-```sh
-pip install "friendli-client[mllib]"
-```
+> [!NOTE]
+> If you have a Hugging Face checkpoint and want to convert it to a Friendli-compatible format with applying quantization, you need to install the package with the necessary machine learing library (`mllib`) dependencies.
+> In this case, install the package with the following command:
+>
+> ```sh
+> pip install "friendli-client[mllib]"
+> ```
 
 ## Examples
 

friendli/cli/main.py

@@ -89,7 +89,13 @@ def login(
         json={"username": email, "password": password},
         timeout=DEFAULT_REQ_TIMEOUT,
     )
-    resp = r.json()
+    try:
+        resp = r.json()
+    except requests.exceptions.JSONDecodeError:
+        if r.status_code != 200:
+            secho_error_and_exit(r.content.decode())
+        secho_error_and_exit("Invalid response format.")
+
     if "code" in resp and resp["code"] == "mfa_required":
         mfa_token = resp["mfaToken"]
         client = UserMFAClient()

friendli/sdk/api/base.py

@@ -22,7 +22,7 @@
 from friendli.schema.api.v1.codegen.chat_completions_pb2 import V1ChatCompletionsRequest
 from friendli.schema.api.v1.codegen.completions_pb2 import V1CompletionsRequest
 from friendli.utils.request import DEFAULT_REQ_TIMEOUT
-from friendli.utils.url import get_baseurl
+from friendli.utils.url import get_host
 
 _GenerationLine = TypeVar("_GenerationLine", bound=BaseModel)
 
@@ -95,9 +95,9 @@ def __init__(
             endpoint = deployment["endpoint"]
             if not endpoint:
                 raise NotFoundError("Active endpoint for the deployment is not found.")
-            self._endpoint = httpx.URL(get_baseurl(endpoint)).join(deployment_id)
+            self._host = httpx.URL(get_host(endpoint))
         elif endpoint is not None:
-            self._endpoint = httpx.URL(endpoint)
+            self._host = httpx.URL(endpoint)
 
     def _get_headers(self) -> Dict[str, Any]:
         content_type = (
@@ -147,9 +147,9 @@ def _build_url(self, model: Optional[str] = None) -> httpx.URL:
         if model is not None:
             path = model
         if self._deployment_id is not None:
-            path = os.path.join(path, self._deployment_id)
+            path = self._deployment_id
         path = os.path.join(path, self._api_path)
-        return self._endpoint.join(path)
+        return self._host.join(path)
 
     def _build_data(self, data: dict[str, Any]) -> bytes:
         if self._deployment_id is None:
@@ -183,7 +183,8 @@ def _request(
             response = self._client.send(request=request, stream=stream)
             response.raise_for_status()
         except httpx.HTTPStatusError as exc:
-            raise APIError(repr(response)) from exc
+            resp_content = response.read()
+            raise APIError(resp_content.decode()) from exc
 
         return response
 

friendli/utils/url.py

@@ -11,12 +11,12 @@
 discuss_url = "https://discuss.friendli.ai/"
 
 
-def get_baseurl(url: str) -> str:
-    """Get a base of a URL."""
+def get_host(url: str) -> str:
+    """Get a host part of a URL."""
     parsed_url = urlparse(url)
     scheme = parsed_url.scheme
-    base = parsed_url.netloc
-    return f"{scheme}://{base}/"
+    netloc = parsed_url.netloc
+    return f"{scheme}://{netloc}/"
 
 
 class URLProvider: