majentik
/

MERaLiON-3-10B-RotorQuant-MLX-2bit

Automatic Speech Recognition

kv-cache-quantization

Model card Files Files and versions

MERaLiON-3-10B-RotorQuant-MLX-2bit / decoder_config.json

Commit History

Re-upload with properly quantized decoder (encoder/adaptor bf16)

ebafd04
verified

majentik commited on about 21 hours ago

Add MLX quantized model with KV cache compression

e103c43
verified

majentik commited on 2 days ago