EmbeddingGemma-300M — LiteRT-LM package

.litertlm build of google/embeddinggemma-300m for the LiteRT-LM EmbeddingEngine (v0.17.0+), used by the Ruminote app for on-device semantic search / RAG.

embeddinggemma_300m.litertlm — dynamic int8, 768-dim output (MRL truncation available at runtime via EmbeddingOptions.outputSize), sequence buckets 64/128/256/512.
Structure: split litert-torch export — embedder (token → embedding) + text_encoder (sequence → pooled vector) + SentencePiece tokenizer + embedding metadata, packed with litert-lm pack. The split pipeline is verified bit-exact against the original model.
Query prefix: task: search result | query: {text} · document prefix: title: none | text: {text} (per the EmbeddingGemma model card).

License

Gemma is provided under and subject to the Gemma Terms of Use. By downloading this model you agree to those terms, including the Gemma Prohibited Use Policy. This repository redistributes a converted (quantized, repackaged) derivative of google/embeddinggemma-300m; all rights in the underlying model remain with Google.

Downloads last month: -

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for kasimtmc/embeddinggemma-litertlm

Base model

google/embeddinggemma-300m

Finetuned

(274)

this model