context-parallelism for xformers

mesolitica / context-parallelism

Context Parallelism, support Blockwise Attention, Ring Attention and Tree Attention.

2 stars 2 forks source link

Open kuangdao opened 3 months ago

kuangdao commented 3 months ago

i see the feature of this https://github.com/vllm-project/vllm/issues/7519 . i want to know when will this feature can be try ?

huseinzol05 commented 3 months ago

Yeah im developing it right now, wait ye sir