added ** inside steamer

app.py

@@ -5,6 +5,7 @@ from PIL import Image
 import threading
 import spaces
 import accelerate
+import time
 
 DESCRIPTION = '''
 <div>
@@ -56,14 +57,14 @@ def krypton(input,
         # Image is not defined at all
         gr.Error("Uplaod an image for Krypton to work")
 
-    image = Image.open(image)
-    # image = Image.open(requests.get(url, stream=True).raw)
     prompt = ("<|start_header_id|>user<|end_header_id|>\n\n<image>\n{input['text']}<|eot_id|>"
               "<|start_header_id|>assistant<|end_header_id|>\n\n")
-    inputs = processor(prompt, image, return_tensors='pt').to('cuda', torch.float16)
+    
+    image = Image.open(image)
+    inputs = processor(prompt, image, return_tensors='pt').to(0, torch.float16)
 
     # Streamer
-    streamer = TextIteratorStreamer(processor, timeout=10.0, skip_prompt=True, skip_special_tokens=True)
+    streamer = TextIteratorStreamer(processor, **{"skip_special_tokens": False, "skip_prompt": True})
 
     if temperature == 0.0:
         do_sample = False
@@ -81,10 +82,20 @@ def krypton(input,
     thread = threading.Thread(target=model.generate, kwargs=generation_kwargs)
     thread.start()
 
-    outputs = []
-    for text in streamer:
-        outputs.append(text)
-        yield "".join(outputs)
+    buffer = ""
+    time.sleep(0.5)
+    for new_text in streamer:
+        # find <|eot_id|> and remove it from the new_text
+        if "<|eot_id|>" in new_text:
+            new_text = new_text.split("<|eot_id|>")[0]
+        buffer += new_text
+
+        # generated_text_without_prompt = buffer[len(text_prompt):]
+        generated_text_without_prompt = buffer
+        # print(generated_text_without_prompt)
+        time.sleep(0.06)
+        # print(f"new_text: {generated_text_without_prompt}")
+        yield generated_text_without_prompt
 
 chatbot=gr.Chatbot(height=600, label="Krypt AI")
 chat_input = gr.MultimodalTextbox(interactive=True, file_types=["image"], placeholder="Enter your question or upload an image.", show_label=False)