YoozLabs/gemma-4-E4B-it-qat-lean-4bit-mlx
Text Generation • 7B • Updated • 59
Lean QAT
QAT quality at honest 4-bit size: uniform 4-bit g64 MLX builds and GGUF exports, no silent 8-bit upgrades. Yooz Labs (yooz.live).
-
-
YoozLabs/gemma-4-12B-it-qat-lean-4bit-mlx
Text Generation • 12B • Updated • 88 -
YoozLabs/Qwen3.5-0.8B-qat-lean-4bit-mlx
Text Generation • 0.8B • Updated • 28 -
YoozLabs/Qwen3.5-4B-qat-lean-4bit-mlx
Text Generation • 4B • Updated • 76 -
YoozLabs/Qwen3.5-0.8B-qat-GGUF
Text Generation • 0.8B • Updated • 155 • 1 -
YoozLabs/Qwen3.5-4B-qat-GGUF
Text Generation • 4B • Updated • 163 • 2