Skip to content

[#18156][fix] Carry lora_id and share the cache_salt derivation in the KV-aware router - #18160

Open
oneKn8 wants to merge 7 commits into
NVIDIA:mainfrom
oneKn8:fix-kv-router-salt-lora
Open

oneKn8 wants to merge 7 commits into
NVIDIA:mainfrom
oneKn8:fix-kv-router-salt-lora

[#18156][doc] Say the LoRA id wrap guards the router, not the worker

da00f5a
Select commit
Loading
Failed to load commit list.