MXFP8 Open on Hugging Face
Parameters
7.9B
Weights (est.)
≈ 8.9 GB
Mac memory (est.)
16 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/gemma-4-e4b-it-qat-q4_0-unquantized
- Vendor
- Google · Gemma
- Task
- Any to any
- Quantisation
- MXFP8
- Licence
- apache-2.0
- Downloads
- 33 in the last 30 days · 651 all time · 0 likes
- Published
- 2026-06-05 · updated 2026-06-05 (5mo ago)
- Tags
- gemma4
Other MLX builds of google/gemma-4-e4b-it-qat-q4_0-unquantized
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| gemma-4-E4B-it-qat-mxfp4 | MXFP4 | ≈ 5.1 GB | 8 GB+ | 69 |
| gemma-4-E4B-it-qat-nvfp4 | 4-bit | ≈ 5.1 GB | 8 GB+ | 39 |
| gemma-4-E4B-it-qat-mxfp8 (this one) | MXFP8 | ≈ 8.9 GB | 16 GB+ | 33 |