dayyass/muse-as-service
REST API for sentence tokenization and embedding using Multilingual Universal Sentence Encoder.
Wraps TensorFlow Hub's multilingual MUSE model in a Flask/Gunicorn REST API, supporting 16 languages and offering token-based JWT authentication to isolate a single 280MB model instance across multiple projects. Provides both raw HTTP endpoints and a built-in Python client, with optional GPU acceleration via NVIDIA Container Toolkit and Docker deployment.
No commits in the last 6 months. Available on PyPI.
Stars
51
Forks
5
Language
Python
License
MIT
Category
Last pushed
Sep 05, 2021
Monthly downloads
21
Commits (30d)
0
Dependencies
2
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/embeddings/dayyass/muse-as-service"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
FlagOpen/FlagEmbedding
Retrieval and Retrieval-augmented LLMs
Blaizzy/mlx-embeddings
MLX-Embeddings is the best package for running Vision and Language Embedding models locally on...
qdrant/fastembed
Fast, Accurate, Lightweight Python library to make State of the Art Embedding
Merck/Sapiens
Sapiens is a human antibody language model based on BERT.
amansrivastava17/embedding-as-service
One-Stop Solution to encode sentence to fixed length vectors from various embedding techniques