Update app.py

app.py

@@ -14,12 +14,14 @@ HF_TOKEN = os.getenv("HF_TOKEN")
 if HF_TOKEN:
     login(token=HF_TOKEN)
 
+
 st.set_page_config(
     page_title="Code Assistant",
     page_icon="🧠",
     layout="wide",
 )
 
+
 if "history" not in st.session_state:
     st.session_state.history = []
 
@@ -29,10 +31,6 @@ if "datasets" not in st.session_state:
 
 @st.cache_resource(show_spinner=False)
 def get_inference_client() -> InferenceClient:
-    """
-    Create one reusable inference client.
-    If HF_TOKEN is set, it will be used automatically for authenticated requests.
-    """
     return InferenceClient(
         provider="hf-inference",
         api_key=HF_TOKEN,
@@ -74,7 +72,7 @@ def generate_code(
     )
 
     completion = client.chat.completions.create(
-        model=f"{model_id}:hf-inference",
+        model=model_id,
         messages=messages,
         max_tokens=max_tokens,
         temperature=temperature,
@@ -85,17 +83,15 @@ def generate_code(
 
 with st.sidebar:
     st.header("⚙️ Control Plane")
-
     st.caption("Uses Hugging Face Inference API instead of local model loading.")
 
     model_id = st.selectbox(
         "Model",
         [
-            "Qwen/Qwen2.5-Coder-3B-Instruct",
             "HuggingFaceTB/SmolLM2-1.7B-Instruct",
+            "Qwen/Qwen2.5-Coder-3B-Instruct",
             "microsoft/Phi-3.5-mini-instruct",
         ],
-        help="Choose a small instruction or code-capable model available through hf-inference.",
     )
 
     max_tokens = st.slider(
@@ -158,10 +154,10 @@ with st.sidebar:
         )
 
     if not HF_TOKEN:
-        st.warning("HF_TOKEN is not set. Public-rate or auth failures may occur.")
+        st.warning("HF_TOKEN is not set. Auth or rate-limit issues may occur.")
 
 
-st.title("🧠 Code Assistant")
+st.title("🧠 Uncensored HackerCoder Assistant")
 st.caption("Streamlit + Datasets + Hugging Face Inference API")
 
 prompt = st.text_area(