phoenix-oss/llama-stack
1
0
Fork 0
forked from phoenix-oss/llama-stack-mirror
Code Pull requests Releases Packages 2 Activity Actions

Use GPUs 0 and 1

Browse source
This commit is contained in:
Ashwin Bharambe 2024-11-12 14:21:22 -08:00
parent 1245a625ce
commit e4f14eafe2
1 changed files with 2 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

4
distributions/remote-vllm/compose.yaml
View file

@ -22,7 +22,7 @@ services:
devices: devices:
- nvidia.com/gpu=all - nvidia.com/gpu=all
environment: environment:
- CUDA_VISIBLE_DEVICES=4 - CUDA_VISIBLE_DEVICES=0
- HUGGING_FACE_HUB_TOKEN=$HF_TOKEN - HUGGING_FACE_HUB_TOKEN=$HF_TOKEN
command: > command: >
--gpu-memory-utilization 0.75 --gpu-memory-utilization 0.75
@ -48,7 +48,7 @@ services:
devices: devices:
- nvidia.com/gpu=all - nvidia.com/gpu=all
environment: environment:
- CUDA_VISIBLE_DEVICES=5 - CUDA_VISIBLE_DEVICES=1
- HUGGING_FACE_HUB_TOKEN=$HF_TOKEN - HUGGING_FACE_HUB_TOKEN=$HF_TOKEN
command: > command: >
--gpu-memory-utilization 0.75 --gpu-memory-utilization 0.75