majentik
/

MERaLiON-3-10B-RotorQuant-MLX-8bit

Automatic Speech Recognition

kv-cache-quantization

Model card Files Files and versions

MERaLiON-3-10B-RotorQuant-MLX-8bit / decoder_config.json

Commit History

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

f818655
verified

majentik commited on about 17 hours ago

Add MLX quantized model with KV cache compression

ba8ef4c
verified

majentik commited on 2 days ago