Visual Document Retrieval
Transformers
Safetensors
multilingual
llama_nemotron_vl
feature-extraction
text
image
vidore
rag
multimodal-embedding
multilingual-embedding
Text-to-Visual Document (T→VD) retrieval
custom_code
Instructions to use nvidia/llama-nemotron-colembed-vl-3b-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use nvidia/llama-nemotron-colembed-vl-3b-v2 with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("nvidia/llama-nemotron-colembed-vl-3b-v2", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Avoid deprecated bidirectional mask embed keyword
#2
by oliverholworthy - opened
Call create_bidirectional_mask with config and embeddings positionally while keeping attention_mask named, avoiding dependence on either input_embeds or inputs_embeds keyword spelling.
BoLiu changed pull request status to merged