Update app.py

app.py

@@ -38,7 +38,10 @@ model2 = AutoModelForCausalLM.from_pretrained(
     torch_dtype=torch.bfloat16,
     token=access_token
 )
-
+model3 = AutoModelForCausalLM.from_pretrained(
+    "Qwen/Qwen2-0.5B-Instruct",
+    device_map="auto"
+)
 
 @app.get("/")
 async def read_root():
@@ -113,6 +116,37 @@ async def read_droot():
     print(time_taken)
     return {"Hello": "World!"}
 
+@app.get("/teat")
+async def read_droot():
+    starttime = time.time()
+    messages = [
+        {"role": "system", "content": "You are a helpful assistant, Sia, developed by Sushma. You will response in polity and brief."},
+        {"role": "user", "content": "I'm Alok. Who are you?"},
+        {"role": "assistant", "content": "I am Sia, a small language model created by Sushma."},
+        {"role": "user", "content": "How are you?"}
+    ]
+    text = tokenizer.apply_chat_template(
+        messages,
+        tokenize=False,
+        add_generation_prompt=True
+    )
+    model_inputs = tokenizer([text], return_tensors="pt").to(device)
+        
+    generated_ids = model3.generate(
+        model_inputs.input_ids,
+        max_new_tokens=64
+    )
+    generated_ids = [
+        output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
+    ]
+        
+    response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
+    print(response)
+    end_time = time.time()
+    time_taken = end_time - starttime 
+    print(time_taken)
+    return {"Hello": "World!"}    
+
 @app.get("/text")
 async def read_droot():
     starttime = time.time()