mlx-community/gemma-4-12B-it-OptiQ-4bit

Sync chat template from google/gemma-4-31b-it (Google canonical, published 2026-07-09)

index: register the bf16 vision tower so stock VLM loaders (mlx-vlm, oMLX, LM Studio) can find it

Move OptiQ sidecars under optiq/ so *.safetensors loaders (mlx-vlm, LM Studio) skip them

fix: audio_tower conv weights to MLX channel-last layout (full-VLM loader compat)

docs: OptiQ brand spelling

card: gemma4_unified 12B needs `import optiq` + mlx-lm from main

card: remove em-dashes, ensure funnel

card: add mlx-optiq funnel banner + CTA

Restore vision_config + optiq_vision marker for image input

Add OptIQ vision sidecar (image+text support, v0.2.0)

Add mixed-precision KV-cache recipe (48 layers, 42@4-bit + 6@8-bit, 4.5 avg)

Add gemma-4-12B-it-OptiQ-4bit: mixed-precision MLX quant (Capability 68.23, +6.40 vs uniform 4-bit)

initial commit