Update app.py

app.py

@@ -17,11 +17,11 @@ model = AutoModelForCausalLM.from_pretrained(MODEL_ID, torch_dtype=torch.bfloat1
 def _generate(text: str):
     inputs = tok.apply_chat_template(
         [{"role": "user", "content": f"Summarize in 3 bullets:\n\n{text}"}],
-        return_tensors="pt", add_generation_prompt=True,
+        return_tensors="pt", return_dict=True, add_generation_prompt=True,
     ).to("cuda")
     streamer = TextIteratorStreamer(tok, skip_prompt=True, skip_special_tokens=True)
     Thread(target=model.generate, kwargs=dict(
-        inputs=inputs, streamer=streamer, max_new_tokens=300, do_sample=False,
+        **inputs, streamer=streamer, max_new_tokens=300, do_sample=False,
     )).start()
     return streamer