Feature Extraction
sentence-transformers
PyTorch
Safetensors
Transformers
multilingual
llama_bidirec
text
text-embeddings
retrieval
semantic-search
custom_code
text-embeddings-inference
Instructions to use nvidia/llama-nemotron-embed-1b-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use nvidia/llama-nemotron-embed-1b-v2 with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("nvidia/llama-nemotron-embed-1b-v2", trust_remote_code=True) sentences = [ "The weather is lovely today.", "It's so sunny outside!", "He drove to the stadium." ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] - Transformers
How to use nvidia/llama-nemotron-embed-1b-v2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="nvidia/llama-nemotron-embed-1b-v2", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("nvidia/llama-nemotron-embed-1b-v2", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
Use positional bidirectional mask helper call
#22
by nvidia-oliver-holworthy - opened
Update the custom bidirectional Llama model to call transformers.masking_utils.create_bidirectional_mask with the current inputs_embeds keyword instead of the deprecated input_embeds alias.
nvidia-oliver-holworthy changed pull request title from Use inputs_embeds for bidirectional mask helper to Use positional bidirectional mask helper call
Updated this PR to use positional arguments for create_bidirectional_mask(config, inputs, attention_mask), avoiding dependency on either the deprecated input_embeds keyword or the newer inputs_embeds spelling.
BoLiu changed pull request status to merged