Visual Question Answering
Transformers
Safetensors
English
videollama2_qwen2
text-generation
Audio-visual Question Answering
Audio Question Answering
multimodal large language model
Eval Results
Instructions to use DAMO-NLP-SG/VideoLLaMA2.1-7B-AV with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use DAMO-NLP-SG/VideoLLaMA2.1-7B-AV with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "visual-question-answering" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # pip install "transformers<5.0.0" from transformers import pipeline pipe = pipeline("visual-question-answering", model="DAMO-NLP-SG/VideoLLaMA2.1-7B-AV")# pip install -U transformers accelerate # Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("DAMO-NLP-SG/VideoLLaMA2.1-7B-AV", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Add OmniCompass evaluation results
#5
by VCLab-PolyU - opened
Add the OmniCompass benchmark authors' published results for VideoLLaMA2.1-7B-AV on omnicompass_av. Each value is Overall Cross-Specification Consistency Score on a 0-100 scale; higher is better.
Source: https://omnicompass.netlify.app/#leaderboard
Benchmark: https://huggingface.co/datasets/VCLab-PolyU/OmniCompass
The YAML preserves the published one-decimal scores and records frame counts and source attribution. It adds evaluation metadata only and makes no HF verified claim.