Downloads · 30 days
128
66% of all-time downloads
diffgemma-26b-a4b-it-q4
mmastrac/diffgemma-26b-a4b-it-q4
diffgemma-26b-a4b-it-q4 is a text generation model from mmastrac. Use it when you need the model to write or continue text. The card lists the license as gemma.
A quantized, self-contained .dgq pack of DiffusionGemma 26B-A4B-it (Gemma-4 26B-A4B MoE, discrete block diffusion) for the diffgemma-mps Rust + Metal inference engine on Apple Silicon. Text-only (v1).
All-time downloads
194
Public
Repo size
20.3 GB
Likes
1
Public
Repo makeup
Click a slice to open those files.
.bin20.2 GB · 100%
At a glance
- Task
- Text Generation
- License
- gemma
- Model type
- diffusion_gemma
- Access
- Public
- Created
- Jul 26, 2026
- Updated
- Jul 31, 2026
- SHA
- be312db8
Try a prompt
Base models
Model card
From the Hugging Face model README
DiffGemma 26B-A4B-it — q4 (.dgq)
A quantized, self-contained .dgq pack of
DiffusionGemma 26B-A4B-it
(Gemma-4 26B-A4B MoE, discrete block diffusion) for the
diffgemma-mps Rust + Metal inference
engine on Apple Silicon. Text-only (v1).
Quantization — q4 profile
| Tensor class | Format | Bits/weight |
|---|---|---|
| MoE experts | q4_block (int4 affine, group 32) | 5.0 |
| Attention (q/k/v/o), dense FFN | bf16 | 16 |
| Embeddings, router, norms | bf16 | 16 |
| Self-conditioning MLP | q8_row | 8 |
~19 GiB of weights. Manifest version 1 (affine) — loads on every diffgemma-mps build.
Usage
diffgemma-mps download --repo mmastrac/diffgemma-26b-a4b-it-q4 -o model/diffgemma-26b-a4b-it-q4
diffgemma-mps -m model/diffgemma-26b-a4b-it-q4 chat
Build and run details: github.com/mmastrac/diffgemma.
License
Gemma Terms of Use. Derived from
google/diffusiongemma-26B-A4B-it; use is subject to the Gemma license.
- Task
- Text Generation
- Type
- diffusion_gemma
- License
- gemma
- Created
- Jul 26, 2026
- Updated
- Jul 31, 2026