lint and format

Makefile

@@ -43,7 +43,7 @@ py-upgrade:	## Upgrade project py files with pyupgrade library for python versio
 
 .PHONY: lint
 lint:  ## Lint project code.
-	poetry run ruff check --fix .
+	uv run ruff check --fix .
 
 .PHONY: slim-build
 slim-build: ## with power of docker-slim build smaller and safer images

app/main.py

@@ -7,11 +7,11 @@ from apscheduler.eventbrokers.redis import RedisEventBroker
 from fastapi import Depends, FastAPI
 
 from app.api.health import router as health_router
+from app.api.ml import router as ml_router
 from app.api.nonsense import router as nonsense_router
 from app.api.shakespeare import router as shakespeare_router
 from app.api.stuff import router as stuff_router
 from app.api.user import router as user_router
-from app.api.ml import router as ml_router
 from app.config import settings as global_settings
 from app.database import engine
 from app.redis import get_redis

app/services/llm.py

@@ -1,6 +1,7 @@
+from collections.abc import AsyncGenerator
+
 import httpx
 import orjson
-from typing import AsyncGenerator, Optional
 
 
 class StreamLLMService:
@@ -8,7 +9,7 @@ class StreamLLMService:
         self.base_url = base_url
         self.model = "llama3.2"
 
-    async def stream_chat(self, prompt: str) -> AsyncGenerator[bytes, None]:
+    async def stream_chat(self, prompt: str) -> AsyncGenerator[bytes]:
         """Stream chat completion responses from LLM."""
         # Send the user a message first
         user_msg = {
@@ -47,5 +48,5 @@ class StreamLLMService:
 
 
 # FastAPI dependency
-def get_llm_service(base_url: Optional[str] = None) -> StreamLLMService:
+def get_llm_service(base_url: str | None = None) -> StreamLLMService:
     return StreamLLMService(base_url=base_url or "http://localhost:11434/v1")

tests/chat.py

@@ -2,6 +2,7 @@ import anyio
 import httpx
 import orjson
 
+
 async def chat_with_endpoint():
     async with httpx.AsyncClient() as client:
         while True:
@@ -14,9 +15,9 @@ async def chat_with_endpoint():
             print("\nModel: ", end="", flush=True)
             async with client.stream(
                 "POST",
-                "http://localhost:8000/chat/",
+                "http://0.0.0.0:8080/v1/ml/chat/",
                 data={"prompt": prompt},
-                timeout=60
+                timeout=60,
             ) as response:
                 async for chunk in response.aiter_lines():
                     if chunk:
@@ -26,5 +27,6 @@ async def chat_with_endpoint():
                         except Exception as e:
                             print(f"\nError parsing chunk: {e}")
 
+
 if __name__ == "__main__":
     anyio.run(chat_with_endpoint)