Update app.py

app.py

@@ -36,7 +36,7 @@ llm = Llama.from_pretrained(
   repo_id="openbmb/MiniCPM-Llama3-V-2_5-gguf",
   filename="ggml-model-Q5_K_M.gguf",
   # chat_handler=chat_handler,
-  n_ctx=200000,
+  n_ctx=4096,
 )
 
 def image_to_base64_data_uri(file_path):
@@ -51,7 +51,7 @@ def stream_chat(message, history: list, temperature: float, max_new_tokens: int)
     messages = []
 
     if message["files"]:
-        image = image_to_base64_data_uri(message["files"][-1])
+        image = message["files"][-1]
         messages.append({
             "role": "user", 
             "content": [
@@ -64,7 +64,7 @@ def stream_chat(message, history: list, temperature: float, max_new_tokens: int)
             raise gr.Error("Please upload an image first.")
             image = None
         else:
-            image = image_to_base64_data_uri(history[0][0][0])
+            image = history[0][0][0]
             for prompt, answer in history:
                 if answer is None:
                     messages.extend([{