Update main.py

main.py

@@ -115,7 +115,7 @@ async def chat_completions(request: ChatCompletionRequest):
 
     # The payload for the external API uses our system prompt and the combined user query + search results
     payload = {
-        "model": "meta-llama/llama-3.1-8b-instruct", # The actual model used by the inference API
+        "model": "meta-llama/llama-3.1-8b-instruct/fp-8", # The actual model used by the inference API
         "messages": [
             {"role": "system", "content": SYSTEM_PROMPT},
             {"role": "user", "content": final_prompt},