formatting + removed cpu parralism

Dockerfile

@@ -8,7 +8,6 @@ RUN useradd -m -u 1000 user
 ENV HOME=/home/user
 ENV APP_HOME=/home/user/app
 ENV HF_HOME=/home/user/.hf_home
-ENV OMP_NUM_THREADS=8
 
 # 🔹 Set working directory
 WORKDIR $APP_HOME

app.py

@@ -35,10 +35,12 @@ class PromptRequest(BaseModel):
 class ChatResponse(BaseModel):
     response: str
 
+
 @app.get("/")
 def root():
     return {"message": "MASX OpenChat API is running"}
 
+
 @app.get("/status")
 async def status():
     try:

model_loader.py

@@ -11,8 +11,8 @@ load_dotenv()
 # === High-Precision GGUF Model Configuration ===
 MODEL_REPO = os.getenv("MODEL_REPO", "TheBloke/openchat-3.5-0106-GGUF")
 MODEL_FILE = os.getenv("MODEL_FILE", "openchat-3.5-0106.Q8_0.gguf")
-MODEL_TYPE = os.getenv("MODEL_TYPE", "mistral")   # OpenChat 3.5 is Mistral-compatible
-CTX_LEN = int(os.getenv("CTX_LEN", "8192"))       # Use full 8K context
+MODEL_TYPE = os.getenv("MODEL_TYPE", "mistral")  # OpenChat 3.5 is Mistral-compatible
+CTX_LEN = int(os.getenv("CTX_LEN", "8192"))  # Use full 8K context
 
 # === Load Model ===
 model = AutoModelForCausalLM.from_pretrained(