llava-hf
/

llava-interleave-qwen-0.5b-hf

@@ -119,7 +119,7 @@ When prompting with videos/3D/multi-view input, prompt like following:
 image_tokens = "<image>" * n
 prompt = f"<|im_start|>user {image_tokens}\nWhat are these?|im_end|><|im_start|>assistant"
-# With chat template if you sampled 6 frames you have to have 8 images in one conversation turn
 conversation = [
     {

 image_tokens = "<image>" * n
 prompt = f"<|im_start|>user {image_tokens}\nWhat are these?|im_end|><|im_start|>assistant"
+# With chat template if you sampled 5 frames you have to have 5 images in one conversation turn
 conversation = [
     {