MXFP4 Open on Hugging Face
Parameters
744M
Weights (est.)
≈ 1.1 GB
Mac memory (est.)
8 GB+
Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.
Details
- Base model
- google/embeddinggemma-2 (quantized)
- Vendor
- Google · Gemma
- Task
- Embeddings
- Quantisation
- MXFP4
- Licence
- apache-2.0
- Downloads
- 350 in the last 30 days · 859 all time · 3 likes
- Published
- 2026-10-06 · updated 2026-10-06 (1mo ago)
- Languages
- multilingual
- Tags
- embedding_gemma2, mlx-vlm, embedding, sentence-similarity, multimodal, image-feature-extraction, audio-feature-extraction, video-feature-extraction
Other MLX builds of google/embeddinggemma-2
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| embeddinggemma-2-4bit | 4-bit | ≈ 1.1 GB | 8 GB+ | 925 |
| embeddinggemma-2-mxfp4 (this one) | MXFP4 | ≈ 1.1 GB | 8 GB+ | 350 |
| embeddinggemma-2-nvfp4 | 4-bit | ≈ 1.1 GB | 8 GB+ | 325 |
| embeddinggemma-2-5bit | 5-bit | ≈ 1.1 GB | 8 GB+ | 354 |
| embeddinggemma-2-6bit | 6-bit | ≈ 1.2 GB | 8 GB+ | 321 |
| embeddinggemma-2-8bit | 8-bit | ≈ 1.2 GB | 8 GB+ | 5.8K |
| embeddinggemma-2-mxfp8 | MXFP8 | ≈ 1.2 GB | 8 GB+ | 717 |
| embeddinggemma-2-bf16 | bf16 | ≈ 1.5 GB | 8 GB+ | 5.6K |