llama-cpp-docker/docker-compose.yml
2026-03-09 14:30:33 +01:00

20 lines
444 B
YAML

services:
llm:
build: .
container_name: server
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: all
capabilities: [gpu]
restart: unless-stopped
environment:
- NVIDIA_VISIBLE_DEVICES=all
volumes:
- ./models:/models
ports:
- 3000:8080
command: --models-preset /models/model-presets.ini --fit on --fit-target 1024