Files
GenAIExamples/ChatQnA/deployment/nginx
lvliang-intel fabff168ff add initial examples
Signed-off-by: lvliang-intel <liang1.lv@intel.com>
2024-03-21 10:17:09 +08:00
..
2024-03-21 10:17:09 +08:00
2024-03-21 10:17:09 +08:00
2024-03-21 10:17:09 +08:00
2024-03-21 10:17:09 +08:00

Launch 8 models on 8 separate Gaudi2 cards:

Add HuggingFace access token in .env
Optinally change model name and linked volume direcotry to store downloaded model

Run the following command in your terminal to launch nginx load balancer and 8 instances of tgi_gaudi containers (one for each Gaudi card):

docker compose up -f docker-compose.yml -d