Standardized Serverless ML Inference Platform on Kubernetes
APACHE-2.0 License
Bot releases are visible (Hide)
certificate
to ca_bundle
in json style of s3 storageSecret by @Jooho in https://github.com/kserve/kserve/pull/3463
Full Changelog: https://github.com/kserve/kserve/compare/v0.11.1...v0.12.0
Published by yuzisun 9 months ago
Full Changelog: https://github.com/kserve/kserve/compare/v0.12.0-rc0...v0.12.0-rc1
Published by yuzisun 10 months ago
Full Changelog: https://github.com/kserve/kserve/compare/v0.11.1...v0.12.0-rc0
Published by yuzisun 11 months ago
Full Changelog: https://github.com/kserve/kserve/compare/v0.11.1...v0.11.2
Published by yuzisun about 1 year ago
no space left on device
error in predictor-runtime-build
by @ckadner in https://github.com/kserve/kserve/pull/3049
Full Changelog: https://github.com/kserve/kserve/compare/v0.11.0...v0.11.1
Published by yuzisun about 1 year ago
kserve 0.11.0 now uses poetry for dependency management, cloud storage dependencies are now made optional and you can run pip install kserve[storage]
to install those dependencies.
content-type
header to application/json
for the server to recognize and decode the json
type payload.Full Changelog: https://github.com/kserve/kserve/compare/v0.10.0...v0.11.0
Published by yuzisun over 1 year ago
Full Changelog: https://github.com/kserve/kserve/compare/v0.11.0-rc0...v0.11.0-rc1
Published by yuzisun over 1 year ago
Full Changelog: https://github.com/kserve/kserve/compare/v0.10.1...v0.10.2
Published by yuzisun over 1 year ago
quick_install.sh
file by @nagar-ajay in https://github.com/kserve/kserve/pull/2662
Full Changelog: https://github.com/kserve/kserve/compare/v0.10.0...v0.11.0-rc0
Published by yuzisun over 1 year ago
quick_install.sh
file by @nagar-ajay in https://github.com/kserve/kserve/pull/2662
Full Changelog: https://github.com/kserve/kserve/compare/v0.10.0...v0.10.1
Published by yuzisun over 1 year ago
Cleanup InferenceService configmap for ML framework related fields by @Suresh-Nakkeran in https://github.com/kserve/kserve/pull/2225
Warning: If you still have inference service predictors which are not mutated by the webook to convert to the new model spec introduced since 0.8, you need to make a simple spec update to have them go through the mutating webhook.
Pass request headers to predict method (#2284) by @andyi2it in https://github.com/kserve/kserve/pull/2360
Warning: the preprocess, predict and postprocess now requires passing the additional argument with headers
.
v0.10.0
by @ckadner in https://github.com/kserve/kserve/pull/2645
Full Changelog: https://github.com/kserve/kserve/compare/v0.9.0...v0.10.0
Published by yuzisun almost 2 years ago
Full Changelog: https://github.com/kserve/kserve/compare/v0.10.0-rc0...v0.10.0-rc1
Published by yuzisun almost 2 years ago
Full Changelog: https://github.com/kserve/kserve/compare/v0.9.0...v0.10.0-rc0
Published by yuzisun about 2 years ago
In this release we are excited to introduce the new InferenceGraph
feature which has long been asked from the community. Also continuing the effort from the last release for unifying the InferenceService API for deploying models on KServe and ModelMesh, ModelMesh is now fully compatible with KServe InferenceService API!
protocolversion
in servingruntime spec by @Suresh-Nakkeran in https://github.com/kserve/kserve/pull/2118
Full Changelog: https://github.com/kserve/kserve/compare/v0.8.0...v0.9.0
Published by yuzisun over 2 years ago
protocolversion
in servingruntime spec by @Suresh-Nakkeran in https://github.com/kserve/kserve/pull/2118
Full Changelog: https://github.com/kserve/kserve/compare/v0.8.0...v0.9.0-rc0
Published by yuzisun over 2 years ago
This release introduces two new CRDs ServingRuntimes and ClusterServingRuntimes with the only difference between these two is that one is namespace-scoped and one is cluster-scoped. A ServingRuntime defines the templates for Pods that can serve one or more particular model formats. Each ServingRuntime defines key information such as the container image of the runtime and a list of the model formats that the runtime supports.
In previous versions of KServe, supported predictor formats and container images were defined in a config map in the control plane namespace. The ServingRuntime CRD should allow for improved flexibility and extensibility for defining or customizing runtimes to how you see fit without having to modify any controller code or any resources in the controller namespace.
Several out-of-the-box ClusterServingRuntimes are provided with KServe so that users can continue to use KServe how they did before without having to define the runtimes themselves.
Full Changelog: https://github.com/kserve/kserve/compare/v0.7.0...v0.8.0
Published by yuzisun over 2 years ago
Full Changelog: https://github.com/kserve/kserve/compare/v0.7.0...v0.8.0-rc0
Published by yuzisun about 3 years ago
What's Changed?
InferenceService
API group is changed from serving.kubeflow.org
to serving.kserve.io
https://github.com/kserve/kserve/issues/1826.kfserving
to kserve
https://github.com/kserve/kserve/issues/1827.🌈 What's New?
🐛 What's Fixed?
Published by yuzisun about 3 years ago
Published by yuzisun over 3 years ago
🌈 What's New?
🐛 What's Fixed?
What's Changed?
Development experience and docs