DAMO-NLP-SG
/

VideoLLaMA3-7B-Image

@@ -1,135 +1,147 @@
----
-library_name: transformers
-tags:
-- multi-modal
-- large-language-model
-- video-language-model
-license: apache-2.0
-datasets:
-- lmms-lab/LLaVA-OneVision-Data
-- allenai/pixmo-docs
-- HuggingFaceM4/Docmatix
-- lmms-lab/LLaVA-Video-178K
-- ShareGPT4Video/ShareGPT4Video
-language:
-- en
-metrics:
-- accuracy
-pipeline_tag: visual-question-answering
-base_model:
-- Qwen/Qwen2.5-7B-Instruct
----
-<p align="center">
-    <img src="https://cdn-uploads.huggingface.co/production/uploads/626938b16f8f86ad21deb989/tt5KYnAUmQlHtfB1-Zisl.png" width="150" style="margin-bottom: 0.2;"/>
-<p>
-<h3 align="center"><a href="https://arxiv.org/abs/2501.13106">VideoLLaMA 3: Frontier Multimodal Foundation Models for Video Understanding</a></h3>
-<h5 align="center"> If you like our project, please give us a star ⭐ on <a href="https://github.com/DAMO-NLP-SG/VideoLLaMA3">Github</a> for the latest update.  </h5>
-## 📰 News
-<!-- * **[2024.01.23]**  👋👋 Update technical report. If you have works closely related to VideoLLaMA3 but not mentioned in the paper, feel free to let us know. -->
-* **[2024.01.24]**  🔥🔥 Online Demo is available: [VideoLLaMA3-Image-7B](https://huggingface.co/spaces/lixin4ever/VideoLLaMA3-Image), [VideoLLaMA3-7B](https://huggingface.co/spaces/lixin4ever/VideoLLaMA3).
-* **[2024.01.22]**  Release models and inference code of VideoLLaMA 3.
-## 🌟 Introduction
-VideoLLaMA 3 represents a state-of-the-art series of multimodal foundation models designed to excel in both image and video understanding tasks. Leveraging advanced architectures, VideoLLaMA 3 demonstrates exceptional capabilities in processing and interpreting visual content across various contexts. These models are specifically designed to address complex multimodal challenges, such as integrating textual and visual information, extracting insights from sequential video data, and performing high-level reasoning over both dynamic and static visual scenes.
-## 🌎 Model Zoo
-| Model                | Base Model   | HF Link                                                      |
-| -------------------- | ------------ | ------------------------------------------------------------ |
-| VideoLLaMA3-7B       | Qwen2.5-7B   | [DAMO-NLP-SG/VideoLLaMA3-7B](https://huggingface.co/DAMO-NLP-SG/VideoLLaMA3-7B) |
-| VideoLLaMA3-2B       | Qwen2.5-1.5B | [DAMO-NLP-SG/VideoLLaMA3-2B](https://huggingface.co/DAMO-NLP-SG/VideoLLaMA3-2B) |
-| VideoLLaMA3-7B-Image (**This Checkpoint**) | Qwen2.5-7B   | [DAMO-NLP-SG/VideoLLaMA3-7B-Image](https://huggingface.co/DAMO-NLP-SG/VideoLLaMA3-7B-Image) |
-| VideoLLaMA3-2B-Image | Qwen2.5-1.5B | [DAMO-NLP-SG/VideoLLaMA3-2B-Image](https://huggingface.co/DAMO-NLP-SG/VideoLLaMA3-2B-Image) |
-We also upload the tuned vision encoder of VideoLLaMA3-7B for wider application:
-| Model                         | Base Model                | HF Link                                                      |
-| ----------------------------- | ------------------------- | ------------------------------------------------------------ |
-| VideoLLaMA3-7B Vision Encoder | siglip-so400m-patch14-384 | [DAMO-NLP-SG/VL3-SigLIP-NaViT](https://huggingface.co/DAMO-NLP-SG/VL3-SigLIP-NaViT) |
-## 🚀 Main Results
-<img width="500" alt="image" src="https://cdn-uploads.huggingface.co/production/uploads/626938b16f8f86ad21deb989/ArHgZAmidn8Qlz8BwOdJI.png">
-* \* denotes the reproduced results.
-## 🤖 Quick Start
-```python
-import torch
-from transformers import AutoModelForCausalLM, AutoProcessor, AutoModel, AutoImageProcessor
-model_name = "DAMO-NLP-SG/VideoLLaMA3-7B-Image"
-model = AutoModelForCausalLM.from_pretrained(
-    model_name,
-    trust_remote_code=True,
-    device_map="auto",
-    torch_dtype=torch.bfloat16,
-    attn_implementation="flash_attention_2",
-)
-processor = AutoProcessor.from_pretrained(model_name, trust_remote_code=True)
-# Image conversation
-conversation = [
-    {
-        "role": "user",
-        "content": [
-            {"type": "image", "image": {"image_path": "https://github.com/DAMO-NLP-SG/VideoLLaMA3/blob/main/assets/sora.png?raw=true"}},
-            {"type": "text", "text": "What is the woman wearing?"},
-        ]
-    }
-]
-inputs = processor(conversation=conversation, return_tensors="pt")
-inputs = {k: v.cuda() if isinstance(v, torch.Tensor) else v for k, v in inputs.items()}
-if "pixel_values" in inputs:
-    inputs["pixel_values"] = inputs["pixel_values"].to(torch.bfloat16)
-output_ids = model.generate(**inputs, max_new_tokens=128)
-response = processor.batch_decode(output_ids, skip_special_tokens=True)[0].strip()
-print(response)
-```
-## Citation
-If you find VideoLLaMA useful for your research and applications, please cite using this BibTeX:
-```bibtex
-@article{damonlpsg2025videollama3,
-  title={VideoLLaMA 3: Frontier Multimodal Foundation Models for Image and Video Understanding},
-  author={Boqiang Zhang, Kehan Li, Zesen Cheng, Zhiqiang Hu, Yuqian Yuan, Guanzheng Chen, Sicong Leng, Yuming Jiang, Hang Zhang, Xin Li, Peng Jin, Wenqi Zhang, Fan Wang, Lidong Bing, Deli Zhao},
-  journal={arXiv preprint arXiv:2501.13106},
-  year={2025},
-  url = {https://arxiv.org/abs/2501.13106}
-}
-@article{damonlpsg2024videollama2,
-  title={VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMs},
-  author={Cheng, Zesen and Leng, Sicong and Zhang, Hang and Xin, Yifei and Li, Xin and Chen, Guanzheng and Zhu, Yongxin and Zhang, Wenqi and Luo, Ziyang and Zhao, Deli and Bing, Lidong},
-  journal={arXiv preprint arXiv:2406.07476},
-  year={2024},
-  url = {https://arxiv.org/abs/2406.07476}
-}
-@article{damonlpsg2023videollama,
-  title = {Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding},
-  author = {Zhang, Hang and Li, Xin and Bing, Lidong},
-  journal = {arXiv preprint arXiv:2306.02858},
-  year = {2023},
-  url = {https://arxiv.org/abs/2306.02858}
-}
 ```

+---
+library_name: transformers
+tags:
+- multi-modal
+- large-language-model
+- video-language-model
+license: apache-2.0
+datasets:
+- lmms-lab/LLaVA-OneVision-Data
+- allenai/pixmo-docs
+- HuggingFaceM4/Docmatix
+- lmms-lab/LLaVA-Video-178K
+- ShareGPT4Video/ShareGPT4Video
+language:
+- zho
+- eng
+- fra
+- spa
+- por
+- deu
+- ita
+- rus
+- jpn
+- kor
+- vie
+- tha
+- ara
+metrics:
+- accuracy
+pipeline_tag: visual-question-answering
+base_model:
+- Qwen/Qwen2.5-7B-Instruct
+---
+<p align="center">
+    <img src="https://cdn-uploads.huggingface.co/production/uploads/626938b16f8f86ad21deb989/tt5KYnAUmQlHtfB1-Zisl.png" width="150" style="margin-bottom: 0.2;"/>
+<p>
+<h3 align="center"><a href="https://arxiv.org/abs/2501.13106">VideoLLaMA 3: Frontier Multimodal Foundation Models for Video Understanding</a></h3>
+<h5 align="center"> If you like our project, please give us a star ⭐ on <a href="https://github.com/DAMO-NLP-SG/VideoLLaMA3">Github</a> for the latest update.  </h5>
+## 📰 News
+<!-- * **[2024.01.23]**  👋👋 Update technical report. If you have works closely related to VideoLLaMA3 but not mentioned in the paper, feel free to let us know. -->
+* **[2024.01.24]**  🔥🔥 Online Demo is available: [VideoLLaMA3-Image-7B](https://huggingface.co/spaces/lixin4ever/VideoLLaMA3-Image), [VideoLLaMA3-7B](https://huggingface.co/spaces/lixin4ever/VideoLLaMA3).
+* **[2024.01.22]**  Release models and inference code of VideoLLaMA 3.
+## 🌟 Introduction
+VideoLLaMA 3 represents a state-of-the-art series of multimodal foundation models designed to excel in both image and video understanding tasks. Leveraging advanced architectures, VideoLLaMA 3 demonstrates exceptional capabilities in processing and interpreting visual content across various contexts. These models are specifically designed to address complex multimodal challenges, such as integrating textual and visual information, extracting insights from sequential video data, and performing high-level reasoning over both dynamic and static visual scenes.
+## 🌎 Model Zoo
+| Model                | Base Model   | HF Link                                                      |
+| -------------------- | ------------ | ------------------------------------------------------------ |
+| VideoLLaMA3-7B       | Qwen2.5-7B   | [DAMO-NLP-SG/VideoLLaMA3-7B](https://huggingface.co/DAMO-NLP-SG/VideoLLaMA3-7B) |
+| VideoLLaMA3-2B       | Qwen2.5-1.5B | [DAMO-NLP-SG/VideoLLaMA3-2B](https://huggingface.co/DAMO-NLP-SG/VideoLLaMA3-2B) |
+| VideoLLaMA3-7B-Image (**This Checkpoint**) | Qwen2.5-7B   | [DAMO-NLP-SG/VideoLLaMA3-7B-Image](https://huggingface.co/DAMO-NLP-SG/VideoLLaMA3-7B-Image) |
+| VideoLLaMA3-2B-Image | Qwen2.5-1.5B | [DAMO-NLP-SG/VideoLLaMA3-2B-Image](https://huggingface.co/DAMO-NLP-SG/VideoLLaMA3-2B-Image) |
+We also upload the tuned vision encoder of VideoLLaMA3-7B for wider application:
+| Model                         | Base Model                | HF Link                                                      |
+| ----------------------------- | ------------------------- | ------------------------------------------------------------ |
+| VideoLLaMA3-7B Vision Encoder | siglip-so400m-patch14-384 | [DAMO-NLP-SG/VL3-SigLIP-NaViT](https://huggingface.co/DAMO-NLP-SG/VL3-SigLIP-NaViT) |
+## 🚀 Main Results
+<img width="500" alt="image" src="https://cdn-uploads.huggingface.co/production/uploads/626938b16f8f86ad21deb989/ArHgZAmidn8Qlz8BwOdJI.png">
+* \* denotes the reproduced results.
+## 🤖 Quick Start
+```python
+import torch
+from transformers import AutoModelForCausalLM, AutoProcessor, AutoModel, AutoImageProcessor
+model_name = "DAMO-NLP-SG/VideoLLaMA3-7B-Image"
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    trust_remote_code=True,
+    device_map="auto",
+    torch_dtype=torch.bfloat16,
+    attn_implementation="flash_attention_2",
+)
+processor = AutoProcessor.from_pretrained(model_name, trust_remote_code=True)
+# Image conversation
+conversation = [
+    {
+        "role": "user",
+        "content": [
+            {"type": "image", "image": {"image_path": "https://github.com/DAMO-NLP-SG/VideoLLaMA3/blob/main/assets/sora.png?raw=true"}},
+            {"type": "text", "text": "What is the woman wearing?"},
+        ]
+    }
+]
+inputs = processor(conversation=conversation, return_tensors="pt")
+inputs = {k: v.cuda() if isinstance(v, torch.Tensor) else v for k, v in inputs.items()}
+if "pixel_values" in inputs:
+    inputs["pixel_values"] = inputs["pixel_values"].to(torch.bfloat16)
+output_ids = model.generate(**inputs, max_new_tokens=128)
+response = processor.batch_decode(output_ids, skip_special_tokens=True)[0].strip()
+print(response)
+```
+## Citation
+If you find VideoLLaMA useful for your research and applications, please cite using this BibTeX:
+```bibtex
+@article{damonlpsg2025videollama3,
+  title={VideoLLaMA 3: Frontier Multimodal Foundation Models for Image and Video Understanding},
+  author={Boqiang Zhang, Kehan Li, Zesen Cheng, Zhiqiang Hu, Yuqian Yuan, Guanzheng Chen, Sicong Leng, Yuming Jiang, Hang Zhang, Xin Li, Peng Jin, Wenqi Zhang, Fan Wang, Lidong Bing, Deli Zhao},
+  journal={arXiv preprint arXiv:2501.13106},
+  year={2025},
+  url = {https://arxiv.org/abs/2501.13106}
+}
+@article{damonlpsg2024videollama2,
+  title={VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMs},
+  author={Cheng, Zesen and Leng, Sicong and Zhang, Hang and Xin, Yifei and Li, Xin and Chen, Guanzheng and Zhu, Yongxin and Zhang, Wenqi and Luo, Ziyang and Zhao, Deli and Bing, Lidong},
+  journal={arXiv preprint arXiv:2406.07476},
+  year={2024},
+  url = {https://arxiv.org/abs/2406.07476}
+}
+@article{damonlpsg2023videollama,
+  title = {Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding},
+  author = {Zhang, Hang and Li, Xin and Bing, Lidong},
+  journal = {arXiv preprint arXiv:2306.02858},
+  year = {2023},
+  url = {https://arxiv.org/abs/2306.02858}
+}
 ```