mlx-community/YandexGPT-5-Lite-8B-pretrain-Q8-mlx

Quantization vs yandex/YandexGPT-5-Lite-8B-pretrain. mlx-community/YandexGPT-5-Lite-8B-pretrain-Q8-mlx is a quantized version of the other (U32/F16).

On the Hub · Downloads on Model Pulse