Lets keep those models loaded for a bit longer.
This commit is contained in:
parent
ba9afaad72
commit
09642a8de9
1 changed files with 2 additions and 0 deletions
|
@ -43,6 +43,8 @@ spec:
|
|||
value: "*"
|
||||
- name: OLLAMA_MODELS
|
||||
value: &modelPath "/models"
|
||||
- name: OLLAMA_KEEP_ALIVE
|
||||
value: "24h"
|
||||
resources:
|
||||
requests:
|
||||
nvidia.com/gpu: 1 # requesting 1 GPU
|
||||
|
|
Loading…
Reference in a new issue