gemma-4-12B-it-qat-assistant-bf16
mlx-community/gemma-4-12B-it-qat-assistant-bf16
Text generation · mlx
bf16 Open on Hugging Face
Parameters
423M
Weights (est.)
≈ 846 MB
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.
Run it
With mlx-lm on an Apple silicon Mac:
pip install mlx-lmmlx_lm.generate --model mlx-community/gemma-4-12B-it-qat-assistant-bf16 --prompt "Hello"
Details
- Base model
- google/gemma-4-12B-it-qat-q4_0-unquantized-assistant (finetune)
- Vendor
- Google · Gemma
- Task
- Text generation
- Quantisation
- bf16
- Licence
- gemma
- Downloads
- 87 in the last 30 days · 1.7K all time · 0 likes
- Published
- 2026-06-05 · updated 2026-06-05 (5mo ago)
- Languages
- en
- Tags
- gemma4_unified_assistant, mlx-vlm, gemma-4, gemma-4-12b, gemma, speculative-decoding, draft-model, bf16
Other MLX builds of google/gemma-4-12B-it-qat-q4_0-unquantized-assistant
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| gemma-4-12B-it-qat-assistant-4bit | 4-bit | ≈ 238 MB | 8 GB+ | 454 |
| gemma-4-12B-it-qat-assistant-mxfp4 | MXFP4 | ≈ 238 MB | 8 GB+ | 57 |
| gemma-4-12B-it-qat-assistant-nvfp4 | 4-bit | ≈ 238 MB | 8 GB+ | 48 |
| gemma-4-12B-it-qat-assistant-5bit | 5-bit | ≈ 291 MB | 8 GB+ | 56 |
| gemma-4-12B-it-qat-assistant-6bit | 6-bit | ≈ 344 MB | 8 GB+ | 53 |
| gemma-4-12B-it-qat-assistant-8bit | 8-bit | ≈ 449 MB | 8 GB+ | 118 |
| gemma-4-12B-it-qat-assistant-mxfp8 | MXFP8 | ≈ 449 MB | 8 GB+ | 45 |
| gemma-4-12B-it-qat-assistant-bf16 (this one) | bf16 | ≈ 846 MB | 8 GB+ | 87 |