mirror of
https://github.com/hanzo-ml/kserve.git
synced 2026-10-30 20:53:49 +00:00
Signed-off-by: Sergey Bekkerman <[email protected]> Signed-off-by: SeanCondon <[email protected]> Co-authored-by: SeanCondon <[email protected]>
48 lines
1.2 KiB
YAML
48 lines
1.2 KiB
YAML
---
|
|
apiVersion: batch/v1
|
|
kind: Job
|
|
metadata:
|
|
name: gpt-oss-20b-init-job
|
|
spec:
|
|
parallelism: 1
|
|
completions: 1
|
|
template:
|
|
spec:
|
|
restartPolicy: Never
|
|
volumes:
|
|
- name: pvc
|
|
persistentVolumeClaim:
|
|
claimName: gpt-oss-20b-pvc
|
|
readOnly: false
|
|
containers:
|
|
- image: kserve/kserve-storage-initializer:v0.17.0
|
|
name: storage-initializer
|
|
args:
|
|
- hf://RedHatAI/gpt-oss-20b
|
|
- /mnt/models
|
|
env:
|
|
- name: HF_HOME
|
|
value: /tmp/hf
|
|
- name: HF_XET_NUM_CONCURRENT_RANGE_GETS
|
|
value: "8"
|
|
- name: HF_XET_HIGH_PERFORMANCE
|
|
value: "True"
|
|
- name: HF_HUB_DISABLE_TELEMETRY
|
|
value: "1"
|
|
- name: HF_TOKEN
|
|
valueFrom:
|
|
secretKeyRef:
|
|
name: hf-token
|
|
key: HF_TOKEN
|
|
resources:
|
|
requests:
|
|
cpu: "2"
|
|
memory: "16Gi"
|
|
limits:
|
|
cpu: "8"
|
|
memory: "32Gi"
|
|
volumeMounts:
|
|
- name: pvc
|
|
mountPath: "/mnt/models"
|
|
readOnly: false
|