Run sentence embeddings with Keras 3

zeromodels/paraphrase-albert-small-v2

Pure-Keras 3 conversion of sentence-transformers/paraphrase-albert-small-v2 for ZeroModels. It encodes text into 768-dimensional vectors using a albert backbone. Its pooling is mean, its embeddings are not normalized, and the intended scoring function is cosine similarity. The conversion preserves the source model's embedding recipe; it does not introduce a new training run.

For details about training and intended use, see the original model card.

✨ Quick start

import os
os.environ["KERAS_BACKEND"] = "torch"  # set before importing Keras / ZeroModels

from keras import ops
from zeromodels.sentence_transformers import (
    ZMSentenceTransformer,
    ZMSentenceTransformerTokenizer,
)

repo = "zeromodels/paraphrase-albert-small-v2"
model = ZMSentenceTransformer.from_weights(repo)
tokenizer = ZMSentenceTransformerTokenizer.from_weights(repo)

texts = ["A person is reading a book.", "Someone reads a novel."]
embeddings = model.encode(texts, tokenizer=tokenizer)
scores = ops.convert_to_numpy(model.similarity(embeddings, embeddings))
print(embeddings.shape)  # (2, 768)
print(scores)

model.similarity() uses cosine similarity, matching the source model. If you need unit-length vectors for a checkpoint that is not normalized, pass normalize=True to model.encode(...); this changes the original embedding recipe and may change retrieval scores.

Model details

All 13 converted variants use from_weights("zeromodels/<variant>"):

Tips

• Set KERAS_BACKEND before importing Keras or ZeroModels. Embedding parity was checked on the Torch backend; full TensorFlow/JAX parity is not claimed.
• Use the tokenizer from the same repo as the model. It preserves the source truncation length of 100 tokens for this checkpoint.
• For exact source loading without these converted Keras weights, use ZMSentenceTransformer.from_weights("hf:sentence-transformers/paraphrase-albert-small-v2").
• The repository contains model.weights.h5, zm_config.json, and tokenizer.json. The pinned source commit is listed above.

Verification and attribution

The conversion was compared with Sentence Transformers at the pinned source revision. Maximum embedding absolute difference: 2.74181e-06; minimum embedding cosine: 0.99999994; maximum similarity-score absolute difference: 2.38419e-07.

Special thanks to the Sentence Transformers team for the original checkpoint. The upstream card declares apache-2.0.

ZeroModels documentation · Source model · This checkpoint