Update app.py

app.py

@@ -4,9 +4,15 @@ from transformers import AutoTokenizer, AutoModelForCausalLM
 
 model_name = "ICEPVP8977/Uncensored_Qwen2.5_Coder_3B"
 
-tokenizer = AutoTokenizer.from_pretrained(model_name)
+tokenizer = AutoTokenizer.from_pretrained(
+    model_name,
+    trust_remote_code=True,
+    use_fast=False
+)
+
 model = AutoModelForCausalLM.from_pretrained(
     model_name,
+    trust_remote_code=True,
     torch_dtype=torch.float32,
     low_cpu_mem_usage=True
 )
@@ -14,36 +20,33 @@ model = AutoModelForCausalLM.from_pretrained(
 model.eval()
 
 def generate(prompt):
-    messages = [
-        {"role": "user", "content": prompt}
-    ]
-
+    messages = [{"role": "user", "content": prompt}]
     text = tokenizer.apply_chat_template(
         messages,
         tokenize=False,
         add_generation_prompt=True
     )
 
-    inputs = tokenizer(text, return_tensors="pt")
+    inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=512)
 
     with torch.no_grad():
         outputs = model.generate(
             **inputs,
             max_new_tokens=256,
+            do_sample=True,
             temperature=0.7,
-            do_sample=True
+            pad_token_id=tokenizer.eos_token_id
         )
 
-    response = tokenizer.decode(
+    result = tokenizer.decode(
         outputs[0][inputs["input_ids"].shape[-1]:],
         skip_special_tokens=True
     )
-
-    return response
+    return result.strip()
 
 gr.Interface(
     fn=generate,
-    inputs=gr.Textbox(lines=4),
-    outputs="text",
-    title="Qwen2.5 Coder 3B"
+    inputs=gr.Textbox(lines=4, label="Tanya apa saja"),
+    outputs=gr.Textbox(label="Jawaban"),
+    title="AI Assistant",
 ).launch()