Update README.md

README.md

@@ -1,6 +1,7 @@
 ---
 datasets:
 - lmms-lab/LLaVA-NeXT-Video-SFT-Data
+- lmms-lab/LLaVA-OneVision-Data
 language:
 - en
 library_name: transformers
@@ -112,6 +113,8 @@ model-index:
       value: 63.3
       name: accuracy
       verified: true
+base_model:
+- lmms-lab/llava-onevision-qwen2-7b-si
 ---
 
 # LLaVA-Video-7B-Qwen2
@@ -127,7 +130,7 @@ model-index:
 
 ## Model Summary
 
-The LLaVA-OneVision models are 7/72B parameter models trained on [LLaVA-Video-178K](https://huggingface.co/datasets/lmms-lab/LLaVA-NeXT-Video-SFT-Data), based on Qwen2 language model with a context window of 32K tokens.
+The LLaVA-Video models are 7/72B parameter models trained on [LLaVA-Video-178K](https://huggingface.co/datasets/lmms-lab/LLaVA-NeXT-Video-SFT-Data), based on Qwen2 language model with a context window of 32K tokens.
 
 - **Repository:** [LLaVA-VL/LLaVA-NeXT](https://github.com/LLaVA-VL/LLaVA-NeXT?tab=readme-ov-file)
 - **Point of Contact:** [Yuanhan Zhang](https://zhangyuanhan-ai.github.io/)
@@ -179,7 +182,7 @@ def load_video(self, video_path, max_frames_num,fps=1,force_sample=False):
     spare_frames = vr.get_batch(frame_idx).asnumpy()
     # import pdb;pdb.set_trace()
     return spare_frames,frame_time,video_time
-pretrained = "lmms-lab/LLaVA-NeXT-Video-7B-Qwen2"
+pretrained = "lmms-lab/LLaVA-Video-7B-Qwen2"
 model_name = "llava_qwen"
 device = "cuda"
 device_map = "auto"