Embeddings and rerankers · Google
EmbeddingGemma 300M
300M embedding model built for on-device use.
Datasheet
Maker | |
|---|---|
Parameters | 303 MVerifiedsafetensors.total from the HF API Source |
Licence | Gemma Terms of UseVerified |
Access | Gated: accept the terms on Hugging Face to download |
Task (HF pipeline) | sentence-similarity |
First published on HF | 2025-07-17 |
Last updated on HF | 2025-09-25 |
Weights in memory | 0.6 GB at 16-bit · 0.2 GB at ~4.85 bitsEstimateParameters × bits ÷ 8. Excludes context cache and runtime overhead; not every model has a 4-bit build. |
Benchmark
Not yet measured
MMTEB (multilingual) mean task score; rerankers: MTEB-R
No comparable public score for this benchmark yet (Model card is gated; not read). We will not estimate one.
The description summarises the model card at https://huggingface.co/google/embeddinggemma-300m. Parameters and licence are read from the Hugging Face API; the weights-in-memory line is our arithmetic, not a measurement.