mirror of
https://github.com/hanzo-ml/kserve.git
synced 2026-10-30 20:53:49 +00:00
0cc7b33d97
Signed-off-by: ZabinskiMichal <[email protected]> Signed-off-by: Michał Żabiński <[email protected]> Signed-off-by: Spolti <[email protected]> Co-authored-by: Spolti <[email protected]>
232 lines
5.9 KiB
YAML
232 lines
5.9 KiB
YAML
kserve:
|
|
version: v0.19.0
|
|
llmisvcConfigs:
|
|
enabled: false
|
|
servingruntime:
|
|
enabled: false
|
|
modelNamePlaceholder: "{{.Name}}"
|
|
tensorflow:
|
|
disabled: false
|
|
image: tensorflow/serving
|
|
tag: 2.6.2
|
|
imagePullSecrets: []
|
|
imagePullPolicy: IfNotPresent
|
|
securityContext:
|
|
runAsUser: 1000 # User is not defined in the Dockerfile, so we need to set it here to run as non-root
|
|
allowPrivilegeEscalation: false
|
|
privileged: false
|
|
runAsNonRoot: true
|
|
capabilities:
|
|
drop:
|
|
- ALL
|
|
mlserver:
|
|
disabled: false
|
|
image: docker.io/seldonio/mlserver
|
|
tag: 1.5.0
|
|
modelClassPlaceholder: "{{.Labels.modelClass}}"
|
|
imagePullSecrets: []
|
|
imagePullPolicy: IfNotPresent
|
|
securityContext:
|
|
allowPrivilegeEscalation: false
|
|
privileged: false
|
|
runAsNonRoot: true
|
|
capabilities:
|
|
drop:
|
|
- ALL
|
|
sklearnserver:
|
|
disabled: false
|
|
image: kserve/sklearnserver
|
|
tag: ""
|
|
imagePullSecrets: []
|
|
imagePullPolicy: IfNotPresent
|
|
securityContext:
|
|
allowPrivilegeEscalation: false
|
|
privileged: false
|
|
runAsNonRoot: true
|
|
capabilities:
|
|
drop:
|
|
- ALL
|
|
xgbserver:
|
|
disabled: false
|
|
image: kserve/xgbserver
|
|
tag: ""
|
|
imagePullSecrets: []
|
|
imagePullPolicy: IfNotPresent
|
|
securityContext:
|
|
allowPrivilegeEscalation: false
|
|
privileged: false
|
|
runAsNonRoot: true
|
|
capabilities:
|
|
drop:
|
|
- ALL
|
|
huggingfaceserver:
|
|
disabled: false
|
|
image: kserve/huggingfaceserver
|
|
tag: ""
|
|
imagePullSecrets: []
|
|
imagePullPolicy: IfNotPresent
|
|
securityContext:
|
|
allowPrivilegeEscalation: false
|
|
privileged: false
|
|
runAsNonRoot: true
|
|
capabilities:
|
|
drop:
|
|
- ALL
|
|
lmcacheUseExperimental: "True"
|
|
devShm:
|
|
enabled: false
|
|
sizeLimit: ""
|
|
hostIPC:
|
|
enabled: false
|
|
huggingfaceserver_multinode:
|
|
disabled: false
|
|
imagePullSecrets: []
|
|
securityContext:
|
|
allowPrivilegeEscalation: false
|
|
privileged: false
|
|
runAsNonRoot: true
|
|
capabilities:
|
|
drop:
|
|
- ALL
|
|
shm:
|
|
enabled: true
|
|
sizeLimit: "3Gi"
|
|
tritonserver:
|
|
disabled: false
|
|
image: nvcr.io/nvidia/tritonserver
|
|
tag: 23.05-py3
|
|
imagePullSecrets: []
|
|
imagePullPolicy: IfNotPresent
|
|
securityContext:
|
|
runAsUser: 1000 # https://docs.nvidia.com/deeplearning/triton-inference-server/user-guide/docs/customization_guide/deploy.html#run-as-a-non-root-user
|
|
allowPrivilegeEscalation: false
|
|
privileged: false
|
|
runAsNonRoot: true
|
|
capabilities:
|
|
drop:
|
|
- ALL
|
|
pmmlserver:
|
|
disabled: false
|
|
image: kserve/pmmlserver
|
|
tag: ""
|
|
imagePullSecrets: []
|
|
imagePullPolicy: IfNotPresent
|
|
securityContext:
|
|
allowPrivilegeEscalation: false
|
|
privileged: false
|
|
runAsNonRoot: true
|
|
capabilities:
|
|
drop:
|
|
- ALL
|
|
predictiveserver:
|
|
disabled: false
|
|
image: kserve/predictiveserver
|
|
tag: ""
|
|
imagePullSecrets: []
|
|
imagePullPolicy: IfNotPresent
|
|
securityContext:
|
|
allowPrivilegeEscalation: false
|
|
privileged: false
|
|
runAsNonRoot: true
|
|
capabilities:
|
|
drop:
|
|
- ALL
|
|
paddleserver:
|
|
disabled: false
|
|
image: kserve/paddleserver
|
|
tag: ""
|
|
imagePullSecrets: []
|
|
imagePullPolicy: IfNotPresent
|
|
securityContext:
|
|
allowPrivilegeEscalation: false
|
|
privileged: false
|
|
runAsNonRoot: true
|
|
capabilities:
|
|
drop:
|
|
- ALL
|
|
lgbserver:
|
|
disabled: false
|
|
image: kserve/lgbserver
|
|
tag: ""
|
|
imagePullSecrets: []
|
|
imagePullPolicy: IfNotPresent
|
|
securityContext:
|
|
allowPrivilegeEscalation: false
|
|
privileged: false
|
|
runAsNonRoot: true
|
|
capabilities:
|
|
drop:
|
|
- ALL
|
|
autogluonserver:
|
|
disabled: false
|
|
image: kserve/autogluonserver
|
|
tag: ""
|
|
imagePullSecrets: []
|
|
imagePullPolicy: IfNotPresent
|
|
securityContext:
|
|
allowPrivilegeEscalation: false
|
|
privileged: false
|
|
runAsNonRoot: true
|
|
capabilities:
|
|
drop:
|
|
- ALL
|
|
torchserve:
|
|
disabled: false
|
|
image: pytorch/torchserve-kfs
|
|
tag: 0.9.0
|
|
serviceEnvelopePlaceholder: "{{.Labels.serviceEnvelope}}"
|
|
imagePullSecrets: []
|
|
imagePullPolicy: IfNotPresent
|
|
securityContext:
|
|
runAsUser: 1000 # User ID is not defined in the Dockerfile, so we need to set it here to run as non-root
|
|
allowPrivilegeEscalation: false
|
|
privileged: false
|
|
runAsNonRoot: true
|
|
capabilities:
|
|
drop:
|
|
- ALL
|
|
vllmserver:
|
|
disabled: false
|
|
image: vllm/vllm-openai
|
|
tag: latest
|
|
lmcacheUseExperimental: "True"
|
|
imagePullSecrets: []
|
|
imagePullPolicy: IfNotPresent
|
|
securityContext:
|
|
allowPrivilegeEscalation: false
|
|
privileged: false
|
|
capabilities:
|
|
drop:
|
|
- ALL
|
|
devShm:
|
|
enabled: false
|
|
sizeLimit: ""
|
|
hostIPC:
|
|
enabled: false
|
|
art:
|
|
image: kserve/art-explainer
|
|
defaultVersion: ""
|
|
imagePullSecrets: []
|
|
security:
|
|
autoMountServiceAccountToken: true
|
|
inferenceservice:
|
|
resources:
|
|
limits:
|
|
cpu: "1"
|
|
memory: "2Gi"
|
|
requests:
|
|
cpu: "1"
|
|
memory: "2Gi"
|
|
opentelemetryCollector:
|
|
scrapeInterval: "5s"
|
|
metricReceiverEndpoint: "keda-otel-scaler.keda.svc:4317"
|
|
metricScalerEndpoint: "keda-otel-scaler.keda.svc:4318"
|
|
resource:
|
|
cpuLimit: "1"
|
|
memoryLimit: "2Gi"
|
|
cpuRequest: "200m"
|
|
memoryRequest: "512Mi"
|
|
autoscaler:
|
|
scaleUpStabilizationWindowSeconds: "0"
|
|
scaleDownStabilizationWindowSeconds: "300"
|