Add docker olama-mini for total local model
This commit is contained in:
parent
f668edf35e
commit
d3dfeb3119
1 changed files with 26 additions and 0 deletions
26
docker_compose_miniollama.yaml
Normal file
26
docker_compose_miniollama.yaml
Normal file
|
|
@ -0,0 +1,26 @@
|
||||||
|
services:
|
||||||
|
ollama:
|
||||||
|
image: ollama/ollama:latest
|
||||||
|
container_name: ollama-mini
|
||||||
|
restart: always
|
||||||
|
ports:
|
||||||
|
- "11434:11434"
|
||||||
|
volumes:
|
||||||
|
- ollama_data:/root/.ollama
|
||||||
|
environment:
|
||||||
|
- OLLAMA_KEEP_ALIVE=5m # Unloads model after 5 mins to save RAM
|
||||||
|
- OLLAMA_NUM_PARALLEL=1 # Limits to 1 user at a time for speed
|
||||||
|
deploy:
|
||||||
|
resources:
|
||||||
|
limits:
|
||||||
|
memory: 4G # Change this to 50% of your Mini PC's total RAM
|
||||||
|
cpus: '0.50' # Limit to 50% of CPU
|
||||||
|
|
||||||
|
|
||||||
|
--------------------------------------------
|
||||||
|
Module to pull :configs:
|
||||||
|
command : docker exec mini-ollama ollama pull qwen3-coder:480b-cloud
|
||||||
|
- Cloud:
|
||||||
|
qwen3-coder:480b-cloud - for coding tasks, optimized for speed and accuracy
|
||||||
|
kimi-k2:1t-cloud - for orchestration and general tasks, optimized for versatility
|
||||||
|
ministral-3/glm-4.7-flash - for tool processing
|
||||||
Loading…
Add table
Reference in a new issue