llama : fix KV cache quantization for hybrid Mamba/attention models#1548
Closed
jnovy wants to merge 1 commit intoikawrakow:mainfrom
Closed
llama : fix KV cache quantization for hybrid Mamba/attention models#1548jnovy wants to merge 1 commit intoikawrakow:mainfrom
jnovy wants to merge 1 commit intoikawrakow:mainfrom
