diff --git a/stacks/llm/docker-compose.yml b/stacks/llm/docker-compose.yml index f1123e5..6b17f6d 100644 --- a/stacks/llm/docker-compose.yml +++ b/stacks/llm/docker-compose.yml @@ -36,7 +36,8 @@ services: - "99" - "--ctx-size" - "32768" - - "-fa" + - "--flash-attn" + - "on" - "--cache-type-k" - "q8_0" - "--cache-type-v"