EmbeddingGemma (LiteRT): Memoyad mirror

These are unmodified copies of EmbeddingGemma LiteRT files. They are hosted here so the Memoyad app always has a stable download source.

EmbeddingGemma 2 Text Vision 440M

Source: litert-community/embeddinggemma-2-text-vision-440m-litert-lm at commit 538e6574db281b4fe17993a3212f10c73bd678c6. Main model card: google/embeddinggemma-2.

One file embeds images and text into the same 768-dimensional space. It runs with LiteRT-LM on the CPU and GPU.

File Size (bytes) SHA-256
embeddinggemma-2-text-vision-440m.litertlm 387710976 92dcbea108899e5d6e30d919b0744f90d9967e80c67a4ab5503ac16d54f62eb0

This is the generic CPU/GPU build. The source repository also has builds for specific phone chips (NPU); those are not mirrored here.

EmbeddingGemma 300M

Source: litert-community/embeddinggemma-300m

File Size (bytes) SHA-256
embeddinggemma-300M_seq256_mixed-precision.tflite 179131736 37115ef7bff76cd37dd86abe503ff511b1032bf85fc624a85c49c84899e92bc5
sentencepiece_for_embeddinggemma.model 4683319 d6daa52d93d7aad10e8388bd526c4e501d914b47177398d1d9621f1fe48438c7

How the app uses these files

No weights, graphs or tokenizer data were changed. Each file is byte-identical to its source. The app pins each download to a commit of this repository and checks the SHA-256 above. Adding files here, or editing this page, never changes what an installed app downloads.

Licenses

The two model families are under different licenses.

EmbeddingGemma 2 (embeddinggemma-2-text-vision-440m.litertlm) is licensed under the Apache License, Version 2.0, as published by its sources (Google's license page). A copy of the license is in LICENSE-APACHE-2.0.txt.

EmbeddingGemma 300M (embeddinggemma-300M_seq256_mixed-precision.tflite and sentencepiece_for_embeddinggemma.model): Gemma is provided under and subject to the Gemma Terms of Use found at ai.google.dev/gemma/terms

A copy of the terms is in GEMMA_TERMS_OF_USE.md. Use of this model is subject to the Gemma Prohibited Use Policy, which is part of those terms. By downloading or using these files you agree to them.

Credits

EmbeddingGemma and EmbeddingGemma 2 were developed by Google DeepMind. The LiteRT community (Google AI Edge) converted them to LiteRT. This repository is not affiliated with or endorsed by Google.

Downloads last month
45
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for tibaf/embeddinggemma

Finetuned
(181)
this model