Sentence Similarity
sentence-transformers
Safetensors
embedding_gemma2
embedding
multilingual
turkish
retrieval
audio
image
norovox
Instructions to use emrevrg/Nythos-Embed-Zero with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use emrevrg/Nythos-Embed-Zero with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("emrevrg/Nythos-Embed-Zero") sentences = [ "That is a happy person", "That is a happy dog", "That is a very happy person", "Today is a sunny day" ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [4, 4] - Notebooks
- Google Colab
- Kaggle
File size: 466 Bytes
e46da39 9d648d4 | 1 2 3 4 | Nythos Embed Zero (Norovox Labs). Derived from google/embeddinggemma-2 (Apache-2.0).
Text path: weight-space mixture (alpha = 0.7) of EmbeddingGemma 2 and a Norovox-trained soup (karisim8); training used verified distillation from EmbeddingGemma 2 and Qwen3-Embedding-8B (Apache-2.0) on the Norovox pair corpus and our own synthetic data.
Audio and image modality layers retrained by Norovox Labs; the rest of the vision and audio towers come from EmbeddingGemma 2.
|