Update api_clients.py

api_clients.py

@@ -21,26 +21,15 @@ from web_extraction import extract_website_content, enhance_query_with_search
 
 # HF Inference Client
 HF_TOKEN = os.getenv('HF_TOKEN')
-GROQ_API_KEY = os.getenv('GROQ_API_KEY')
-FIREWORKS_API_KEY = os.getenv('FIREWORKS_API_KEY')
 
 def get_inference_client(model_id):
-    """Return an InferenceClient configured for Hugging Face, Groq, or Fireworks AI."""
-    if model_id == "moonshotai/Kimi-K2-Instruct":
-        return InferenceClient(
-            base_url="https://api.groq.com/openai/v1",
-            api_key=GROQ_API_KEY
-        )
-    elif model_id.startswith("fireworks/"):
-        return InferenceClient(
-            base_url="https://api.fireworks.ai/inference/v1",
-            api_key=FIREWORKS_API_KEY
-        )
-    else:
-        return InferenceClient(
-            model=model_id,
-            api_key=HF_TOKEN
-        )
+    """Return an InferenceClient with provider based on model_id."""
+    provider = "groq" if model_id == "moonshotai/Kimi-K2-Instruct" else "auto"
+    return InferenceClient(
+        provider=provider,
+        api_key=HF_TOKEN,
+        bill_to="huggingface"
+    )
 
 # Tavily Search Client
 TAVILY_API_KEY = os.getenv('TAVILY_API_KEY')
@@ -52,7 +41,7 @@ if TAVILY_API_KEY:
         print(f"Failed to initialize Tavily client: {e}")
         tavily_client = None
 
-def generation_code(query: Optional[str], image: Optional[gr.Image], file: Optional[str], website_url: Optional[str], _setting: Dict[str, str], _history: Optional[List[Tuple[str, str]]], _current_model: Dict, enable_search: bool = False, language: str = "html"):
+async def generation_code(query: Optional[str], image: Optional[gr.Image], file: Optional[str], website_url: Optional[str], _setting: Dict[str, str], _history: Optional[List[Tuple[str, str]]], _current_model: Dict, enable_search: bool = False, language: str = "html"):
     if query is None:
         query = ''
     if _history is None:
@@ -117,7 +106,7 @@ This will help me create a better design for you."""
         messages.append({'role': 'user', 'content': enhanced_query})
     try:
         completion = client.chat.completions.create(
-            model=_current_model["id"],
+            model=_current_model["id"], # Corrected this line
             messages=messages,
             stream=True,
             max_tokens=5000