[!IMPORTANT] This quantization was uploaded on 2026-06-06 and replaces a previous version. If you downloaded this model before this date, please re-download for the updated weights.

gemma-4-12B-it-assistant-oQ8

This model was quantized using oQ (oMLX v0.4.2rc1) mixed-precision quantization.

Quantization details

• Model type: gemma4_unified_assistant
• Bits: 8
• Group size: 64
• Format: MLX safetensors