gemma-4-12B-it-qat-oQ4e-fp16-mtp
This model was quantized using oQ (oMLX v0.6.3rc1) mixed-precision quantization.
google/gemma-4-12B-it-qat-q4_0-unquantized-assistant MTP head was merged into google/gemma-4-12B-it-qat-q4_0-unquantized.
It is optimized for Mac M1/M2 (fp16).
Quantization details
- Model type: gemma4_unified
- Bits: 4
- Group size: 64
- Format: MLX safetensors
- Non-quant weight dtype: FP16
- Downloads last month
- 43
Model size
2B params
Tensor type
F16
·
U32 ·
F32 ·
BF16 ·
Hardware compatibility
Log In to add your hardware
4-bit
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for Grunzig/gemma-4-12B-it-qat-oQ4e-fp16-mtp
Base model
google/gemma-4-12B Finetuned
google/gemma-4-12B-it