Instructions to use taskyon/static-similarity-mrl-multilingual-v1-d256-int8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use taskyon/static-similarity-mrl-multilingual-v1-d256-int8 with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("taskyon/static-similarity-mrl-multilingual-v1-d256-int8") 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
Taskyon static multilingual embedding
This is a derivative of sentence-transformers/static-similarity-mrl-multilingual-v1, published by the Sentence Transformers project under Apache-2.0.
Provenance
- Upstream model:
sentence-transformers/static-similarity-mrl-multilingual-v1 - Upstream revision:
7264ea07c5365a11d7e6d87dbb6195889a13054f - Transformation: truncate the Matryoshka embedding from 1024 to 256 dimensions, then apply symmetric per-row INT8 quantization
- Builder: Taskyon's
scripts/build-static-embedding-model.mjs - Intended integration: Taskyon local semantic search
The original model and this derivative are distributed under the Apache License 2.0. The bundled LICENSE is copied from the Apache Software Foundation's canonical license. The model weights were modified; the original model authors do not endorse this derivative.
- Downloads last month
- 41
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support