Update app.py

app.py

@@ -52,13 +52,13 @@ footer {
 }
 """
 
-MODEL_ID = "Qwen/Qwen2-1.5B-Instruct"
+MODEL_ID = "01-ai/Yi-1.5-6B-Chat"
 
 model = AutoModelForCausalLM.from_pretrained(
     MODEL_ID,
     torch_dtype=torch.float16,
     device_map="auto"
-)
+).eval()
 
 tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
 
@@ -130,27 +130,24 @@ async def gen_show(script):
 
 @spaces.GPU
 def generator(messages):
-    text = tokenizer.apply_chat_template(
-        messages,
-        tokenize=False,
-        add_generation_prompt=True
+    input_ids = tokenizer.apply_chat_template(
+        conversation=messages,
+        add_generation_prompt=True,
+        tokenize=True,
+        return_tensors='pt'
     )
-    model_inputs = tokenizer([text], return_tensors="pt").to(0)
     
-    generated_ids = model.generate(
-        model_inputs.input_ids, 
+    output_ids = model.generate(
+        input_ids.to('cuda'), 
+        eos_token_id=tokenizer.eos_token_id,
         max_new_tokens=4096,
         temperature=0.5,
         repetition_penalty=1.2,
         )
     
-    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)
-    return response
+    results = tokenizer.decode(output_ids[0][input_ids.shape[1]:], skip_special_tokens=True)
+    print(results)
+    return results
 
 def extract_content(text):
     """Extracts the JSON content from the given text."""