Inference register fix

examples/agents/inflation.py

@@ -14,10 +14,9 @@
 from llama_stack_client.types import Attachment, SamplingParams, UserMessage
 from llama_stack_client.types.agent_create_params import *  # noqa: F403
 from common.client_utils import *  # noqa: F403
+from examples.agents.multi_turn import execute_turns, prompt_to_turn
 from termcolor import cprint
 
-from .multi_turn import execute_turns, prompt_to_turn
-
 
 async def run_main(host: str, port: int, disable_safety: bool = False):
     api_keys = load_api_keys_from_env()

examples/inference/client.py

@@ -19,9 +19,20 @@ async def run_main(host: str, port: int, stream: bool = True):
         base_url=f"http://{host}:{port}",
     )
 
-    message = UserMessage(
-        content="hello world, write me a 2 sentence poem about the moon", role="user"
+    client.models.register(
+        model={
+            "identifier": "Llama3.1-8B-Instruct",
+            "llama_model": "Llama3.1-8B-Instruct",
+            "provider_id": "meta-reference-0",
+            "metadata": {},
+        }
     )
+
+    message = {
+        "role": "user",
+        "content": "hello world, write me a 2 sentence poem about the moon",
+    }
+
     cprint(f"User>{message.content}", "green")
     response = client.inference.chat_completion(
         messages=[message],

examples/inference/client_with_vision.py

@@ -32,13 +32,20 @@ async def run_main(host: str, port: int, stream: bool = True):
 
     data_url = f"data:{mime_type};base64,{encoded_string}"
 
-    message = UserMessage(
-        role="user",
-        content=[
-            {"image": {"uri": data_url}},
-            "Describe what is in this image.",
-        ],
+    client.models.register(
+        model={
+            "identifier": "Llama3.2-11B-Vision-Instruct",
+            "llama_model": "Llama3.2-11B-Vision-Instruct",
+            "provider_id": "meta-reference-0",
+            "metadata": {},
+        }
     )
+
+    message = {
+        "role": "user",
+        "content": [{"image": {"uri": data_url}}, "Describe what is in this image."],
+    }
+
     cprint(f"User>{message.content}", "green")
     response = client.inference.chat_completion(
         messages=[message],