Update llama_api.py

llama_api.py

@@ -53,7 +53,7 @@ client = OpenAI(
 # 🧠 Non-streaming function
 def ask_ollama(prompt: str):
     free_models = [
-        "openai/gpt-oss-20b:free"
+        "meta-llama/llama-3.1-405b-instruct:free"
     ]
 
     last_error = None
@@ -94,7 +94,7 @@ def ask_ollama(prompt: str):
 
 
 
-def ask_ollama_stream(prompt: str, model_name: str = "openai/gpt-oss-20b:free"):
+def ask_ollama_stream(prompt: str, model_name: str = "meta-llama/llama-3.1-405b-instruct:free"):
     """
     Streams response token-by-token for real-time output.
     Includes detailed logging for debugging.