EmbeddingGemma 2 路 text encoder only 路 8-bit MLX
The text encoder of mlx-community/embeddinggemma-2-8bit
(revision 7505ef2f8ddef45efef6d060865f27989b3c9cec), on its own: about 290 MB instead of 1.2 GB.
It is what Hippo Dash downloads to search your memory by meaning,
running on your Mac with MLX.
Modified: only the language_model.* tensors were kept, byte for byte, re-packed into
model.safetensors; config.json has the image, audio and video encoders removed. Nothing was
re-quantized or retrained. Vectors are 768 floats (Matryoshka: 512, 256 and 128 also work).
License
Gemma is provided under and subject to the Gemma Terms of Use found at ai.google.dev/gemma/terms. Use of this model must also follow the Gemma Prohibited Use Policy.
- Downloads last month
- 54
Model size
0.3B params
Tensor type
BF16
路
U32 路
Hardware compatibility
Log In to add your hardware
8-bit
Model tree for Sagardxd/hippo-embeddinggemma-2-text-8bit
Base model
google/embeddinggemma-2 Quantized
mlx-community/embeddinggemma-2-8bit