Update app.py

app.py

@@ -70,9 +70,11 @@ def run_deepseek(req1, req2, prompt_type):
 
 def run_llama(req1, req2, prompt_type):
     model_name = "meta-llama/Llama-3.1-8B-Instruct"
-    tokenizer = AutoTokenizer.from_pretrained(model_name)
+    hf_token = os.getenv("LLAMA_HF_TOKEN")
+    tokenizer = AutoTokenizer.from_pretrained(model_name, token=hf_token)
     model = AutoModelForCausalLM.from_pretrained(
         model_name,
+        token=hf_token,
         dtype=torch.bfloat16,
         device_map="auto"
     )
@@ -81,8 +83,9 @@ def run_llama(req1, req2, prompt_type):
     outputs = model.generate(inputs.input_ids, max_new_tokens=256)
     return tokenizer.decode(outputs[0], skip_special_tokens=True)
 
+
 def run_fanar(req1, req2, prompt_type):
-    client = OpenAI(base_url="https://api.fanar.qa/v1", api_key="NO8LjYlo9DSZLeL3Yc2dduMCZfQIBCqf")
+    client = OpenAI(base_url="https://api.fanar.qa/v1", api_key=os.getenv("FANAR_API"))
     prompt = build_prompt(req1, req2, prompt_type)
     response = client.chat.completions.create(
         model="Fanar",