From 238cc311f1071509c7642eef4b23b32f79fb9d4e Mon Sep 17 00:00:00 2001 From: OlivierDehaene <23298448+OlivierDehaene@users.noreply.github.com> Date: Thu, 21 Dec 2023 15:46:07 +0100 Subject: [PATCH] back to v2 by def --- server/text_generation_server/utils/layers.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/server/text_generation_server/utils/layers.py b/server/text_generation_server/utils/layers.py index 87d0b6f6..011a9382 100644 --- a/server/text_generation_server/utils/layers.py +++ b/server/text_generation_server/utils/layers.py @@ -33,7 +33,7 @@ except Exception: HAS_EXLLAMA = False CAN_EXLLAMA = major >= 8 -V2 = os.getenv("EXLLAMA_VERSION", "1") == "2" +V2 = os.getenv("EXLLAMA_VERSION", "2") == "2" if V2 and int(os.getenv("WORLD_SIZE", "1")) > 1: logger.warning( "Disabling exllama v2 and using v1 instead because there are issues when sharding"