From 25d831f091affacf104983f8d013be93ee794d76 Mon Sep 17 00:00:00 2001 From: Andy Bunce Date: Sun, 27 Apr 2025 16:53:14 +0100 Subject: [PATCH] [mod] ollama --- ollama/docker-compose.yml | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/ollama/docker-compose.yml b/ollama/docker-compose.yml index da043bf..eac75a7 100644 --- a/ollama/docker-compose.yml +++ b/ollama/docker-compose.yml @@ -12,10 +12,11 @@ services: environment: - OLLAMA_MAX_LOADED_MODELS=1 - OLLAMA_NO_CUDA=1 + - OLLAMA_FLASH_ATTENTION=1 deploy: resources: limits: - cpus: '4' # Limit CPU cores + cpus: 8 # Limit CPU cores memory: 16G # Limit RAM (adjust based on your system) # devices: # - /dev/dri:/dev/dri