Instructions to use minishlab/potion-base-4M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Model2Vec
How to use minishlab/potion-base-4M with Model2Vec:
from model2vec import StaticModel model = StaticModel.from_pretrained("minishlab/potion-base-4M") embeddings = model.encode(["It's dangerous to go alone!", "It's a secret to everybody."]) print(embeddings.shape) - sentence-transformers
How to use minishlab/potion-base-4M with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("minishlab/potion-base-4M") 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] - Notebooks
- Google Colab
- Kaggle
Add full-test ArguAna CPU replication with reproducible traces
#1
by dmytro-nasyrov - opened
Independent CPU replication using the complete registered ArguAna test split (1,406 queries; 8,674 corpus documents), MTEB 2.27.10 and model revision 9b3cff412d30be9ae8603fe10224c224f3401869. Native nDCG@10 is 0.39561; the Hub value is nDCG@10 x 100 = 39.561. The linked Pharos Production report includes the method, pinned environment, original MTEB export, complete saved rankings, per-query metrics and SHA-256 checks. A separate rank calculation reproduces the aggregate. No training, generation, upstream endorsement or verified badge is claimed. This PR adds one result file only.
stephantulkens changed pull request status to closed