Update app.py

app.py

@@ -18,7 +18,7 @@ print(f"[SYSTEM] | Using {DEVICE} type compute device.")
 DEFAULT_INPUT = "Describe in one paragraph."
 MAX_FRAMES = 64
 
-model_name = "openbmb/MiniCPM-o-2_6"
+model_name = "openbmb/MiniCPM-o-2_6-int4"#"openbmb/MiniCPM-o-2_6"
 
 repo = AutoModel.from_pretrained(model_name, trust_remote_code=True, attn_implementation="sdpa", torch_dtype=torch.bfloat16).to(DEVICE)
 tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
@@ -74,7 +74,7 @@ def generate(image, video, audio, instruction=DEFAULT_INPUT, sampling=False, tem
     params = {
         "image": media,
         "msgs": inputs,
-        #"tokenizer": tokenizer,
+        "tokenizer": tokenizer,
         #"processor": processor,
         "sampling": sampling,
         "temperature": temperature,