Instructions to use Hyukkyu/nv-embed-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Hyukkyu/nv-embed-v2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="Hyukkyu/nv-embed-v2", trust_remote_code=True)# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Hyukkyu/nv-embed-v2", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Commit History
-
Fix layer_outputs handling: transformers 5.x returns tensor, not tuple
Verified signature7d27cd8 -
Fix rotary_emb location - use model level attribute
Verified signatureb73d528 -
Fix rotary_emb location - get from layer self_attn
Verified signatureda8a484 -
Add position_embeddings for transformers 5.x decoder layer compatibility
Verified signatureacfcc1b -
Fix get_usable_length -> get_seq_length for transformers 5.x
Verified signatureff33f68 -
Fix from_legacy_cache compatibility for transformers 5.x
Verified signature5285a93
-
Upload migrated model for transformers 5.0+ compatibility
Verified signature005478f -
initial commit
Verified signaturecb0d1e1