First version
This commit is contained in:
@@ -0,0 +1,21 @@
|
|||||||
|
services:
|
||||||
|
vllm:
|
||||||
|
image: vllm/vllm-openai:latest
|
||||||
|
container_name: vllm
|
||||||
|
network_mode: bridge
|
||||||
|
environment:
|
||||||
|
- HF_TOKEN=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
|
||||||
|
ipc: host
|
||||||
|
ports:
|
||||||
|
- "8000:8000"
|
||||||
|
volumes:
|
||||||
|
- /Var/lib/docker/volumes/vllm/huggingface:/root/.cache/huggingface
|
||||||
|
deploy:
|
||||||
|
resources:
|
||||||
|
reservations:
|
||||||
|
devices:
|
||||||
|
- driver: nvidia
|
||||||
|
count: all
|
||||||
|
capabilities: [ gpu ]
|
||||||
|
runtime: nvidia
|
||||||
|
command: --model NVFP4/Qwen3-Coder-30B-A3B-Instruct-FP4
|
||||||
Reference in New Issue
Block a user