homerquan commited on
Commit
5a0729b
·
verified ·
1 Parent(s): d122027

Clarify text, image and audio support in the model card

Browse files
Files changed (2) hide show
  1. README.md +11 -1
  2. SHA256SUMS +1 -1
README.md CHANGED
@@ -8,9 +8,19 @@ tags:
8
  - gguf
9
  - embeddings
10
  - multimodal
 
 
 
 
11
  ---
12
 
13
- # EmbeddingGemma 2 - MirrorNeuronLab Q4_K_M GGUF
 
 
 
 
 
 
14
 
15
  [Google DeepMind](https://huggingface.co/google/embeddinggemma-2) developed the original EmbeddingGemma 2 model. MirrorNeuronLab converted Google's original weights to GGUF and quantized them without additional training.
16
 
 
8
  - gguf
9
  - embeddings
10
  - multimodal
11
+ - vision
12
+ - audio
13
+ - image-feature-extraction
14
+ - audio-feature-extraction
15
  ---
16
 
17
+ # EmbeddingGemma 2 Multimodal GGUF - Q4_K_M Text + Q8_0 Vision/Audio
18
+
19
+ **Text, image, and audio embeddings in a shared 768-dimensional vector space.** This MirrorNeuronLab release pairs a **Q4_K_M text backbone** with a **Q8_0 companion containing both vision and audio encoders**.
20
+
21
+ For image or audio inputs, load both GGUF files. Docker Model Runner automatically downloads the companion when you pull `hf.co/MirrorNeuronLab/embeddinggemma-2-GGUF:Q4_K_M`. The `Q4_K_M` selection identifies the text backbone; the complete release supports text, images, and audio.
22
+
23
+ Image and audio embedding requests were verified through Docker Model Runner on NVIDIA DGX Spark. The model returns embeddings for search and similarity; it does not generate conversational answers or transcriptions.
24
 
25
  [Google DeepMind](https://huggingface.co/google/embeddinggemma-2) developed the original EmbeddingGemma 2 model. MirrorNeuronLab converted Google's original weights to GGUF and quantized them without additional training.
26
 
SHA256SUMS CHANGED
@@ -3,7 +3,7 @@
3
  cfc7749b96f63bd31c3c42b5c471bf756814053e847c10f3eb003417bc523d30 LICENSE
4
  cffa1796862050596e7e071b181883485b8c32de9da54fed01fcf14933806656 PROVENANCE.json
5
  6293911192eb907c3e962e4ec83ea1e095e83d1aa6bfc3c3d7a7ac2f44c5f3a7 PUBLISHED_VALIDATION.json
6
- ac811f096ac49bfedbffe23f31bec20cd1dcd88af069e212460221dd899c4e0b README.md
7
  3a30564979129978d26ad4c1160fab0a04173e76a9fdf1c1f572bab4dc3055e4 SOURCE_SHA256SUMS
8
  1ac5fe21d24f3d310372a52ecd5d91ba7ad002559e53804967b0b7c0c913f09b TIMINGS.json
9
  0a1e1e46a434406b06b5c2b4484cfe05aa5b0f5ef704caaee92e0c50d58a73dc VALIDATION.json
 
3
  cfc7749b96f63bd31c3c42b5c471bf756814053e847c10f3eb003417bc523d30 LICENSE
4
  cffa1796862050596e7e071b181883485b8c32de9da54fed01fcf14933806656 PROVENANCE.json
5
  6293911192eb907c3e962e4ec83ea1e095e83d1aa6bfc3c3d7a7ac2f44c5f3a7 PUBLISHED_VALIDATION.json
6
+ 9a589e59698084274f1456d993fa9b5f30f698bac7980f9873806c1188cd9a76 README.md
7
  3a30564979129978d26ad4c1160fab0a04173e76a9fdf1c1f572bab4dc3055e4 SOURCE_SHA256SUMS
8
  1ac5fe21d24f3d310372a52ecd5d91ba7ad002559e53804967b0b7c0c913f09b TIMINGS.json
9
  0a1e1e46a434406b06b5c2b4484cfe05aa5b0f5ef704caaee92e0c50d58a73dc VALIDATION.json