SeldonIO · tyndria · Jan 26, 2026 · Jan 23, 2026
diff --git a/README.md b/README.md
@@ -68,7 +68,7 @@ We provide optimized model servers for some of the most popular Deep Learning an
 You only have to upload your model binaries into your preferred object store, in this case we have a trained scikit-learn iris model in a Google bucket:
 
 ```console
-gs://seldon-models/v1.19.0-dev/sklearn/iris/model.joblib
+gs://seldon-models/v1.20.0-dev/sklearn/iris/model.joblib
 ```
 
 Create a namespace to run your model in:
@@ -91,7 +91,7 @@ spec:
   predictors:
   - graph:
       implementation: SKLEARN_SERVER
-      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
       name: classifier
     name: default
     replicas: 1

diff --git a/components/alibi-detect-server/Dockerfile b/components/alibi-detect-server/Dockerfile
@@ -5,7 +5,7 @@ FROM ${BASE_IMAGE}:${VERSION} AS base
 ARG VERSION
 LABEL name="Seldon Alibi Detect Server" \
     vendor="Seldon Technologies" \
-    version="1.19.0-dev" \
+    version="1.20.0-dev" \
     release="1" \
     summary="Alibi Detect Server for Seldon Core" \
     description="The Alibi Detect Server provides outlier, drift and adversarial detection services for Seldon Core"

diff --git a/components/alibi-detect-server/pyproject.toml b/components/alibi-detect-server/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "adserver"
-version = "1.19.0-dev"
+version = "1.20.0-dev"
 description = "Model Explanation Server"
 authors = ["Seldon Technologies Ltd. <hello@seldon.io>"]
 license = "Business Source License 1.1"

diff --git a/components/alibi-explain-server/Dockerfile b/components/alibi-explain-server/Dockerfile
@@ -5,7 +5,7 @@ FROM ${BASE_IMAGE}:${VERSION} AS base
 ARG VERSION
 LABEL name="Seldon Alibi Wrapper" \
       vendor="Seldon Technologies" \
-      version="1.19.0-dev" \
+      version="1.20.0-dev" \
       release="1" \
       summary="Alibi Explainer Wrapper for Seldon Core" \
       description="Allows Seldon Core inference models to run with a black box model explanation model from the Alibi:Explain project"

diff --git a/components/alibi-explain-server/Makefile b/components/alibi-explain-server/Makefile
@@ -76,7 +76,7 @@ clean:
 #
 test_models/sklearn/iris:
 	mkdir -p test_models/sklearn/iris
-	gsutil cp -r gs://seldon-models/v1.19.0-dev/sklearn/iris test_models/sklearn
+	gsutil cp -r gs://seldon-models/v1.20.0-dev/sklearn/iris test_models/sklearn
 
 test_models/explainers/anchor_tabular:
 	mkdir -p test_models/explainers/anchor_tabular

diff --git a/components/alibi-explain-server/pyproject.toml b/components/alibi-explain-server/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "alibiexplainer"
-version = "1.19.0-dev"
+version = "1.20.0-dev"
 description = "Model Explanation Server"
 authors = ["Seldon Technologies Ltd. <hello@seldon.io>"]
 license = "Business Source License 1.1"

diff --git a/components/alibi-explain-server/tests/test_anchor_tabular.py b/components/alibi-explain-server/tests/test_anchor_tabular.py
@@ -29,7 +29,7 @@
 from .make_test_models import make_anchor_tabular
 from .utils import SKLearnServer
 
-IRIS_MODEL_URI = "gs://seldon-models/v1.19.0-dev/sklearn/iris/*"
+IRIS_MODEL_URI = "gs://seldon-models/v1.20.0-dev/sklearn/iris/*"
 
 
 def test_anchor_tabular():

diff --git a/components/alibi-explain-server/tests/test_app.py b/components/alibi-explain-server/tests/test_app.py
@@ -9,7 +9,7 @@
 from .make_test_models import make_anchor_tabular
 from .utils import SKLearnServer
 
-IRIS_MODEL_URI = "gs://seldon-models/v1.19.0-dev/sklearn/iris/*"
+IRIS_MODEL_URI = "gs://seldon-models/v1.20.0-dev/sklearn/iris/*"
 
 
 class TestExplainerApp(AsyncHTTPTestCase):

diff --git a/components/drift-detection/cifar10/cifar10_drift.ipynb b/components/drift-detection/cifar10/cifar10_drift.ipynb
@@ -305,7 +305,7 @@
     "        autoscaling.knative.dev/minScale: \"1\"\n",
     "    spec:\n",
     "      containers:\n",
-    "      - image: seldonio/alibi-detect-server:1.19.0-dev\n",
+    "      - image: seldonio/alibi-detect-server:1.20.0-dev\n",
     "        imagePullPolicy: IfNotPresent\n",
     "        args:\n",
     "        - --model_name\n",

diff --git a/components/outlier-detection/cifar10/cifar10_outlier.ipynb b/components/outlier-detection/cifar10/cifar10_outlier.ipynb
@@ -401,7 +401,7 @@
     "        autoscaling.knative.dev/minScale: \"1\"\n",
     "    spec:\n",
     "      containers:\n",
-    "      - image: seldonio/alibi-detect-server:1.19.0-dev\n",
+    "      - image: seldonio/alibi-detect-server:1.20.0-dev\n",
     "        imagePullPolicy: IfNotPresent\n",
     "        args:\n",
     "        - --model_name\n",

diff --git a/components/rclone-storage-initializer/Dockerfile b/components/rclone-storage-initializer/Dockerfile
@@ -8,7 +8,7 @@ RUN microdnf update -y
 
 LABEL name="Rclone Storage Initializer" \
       vendor="Seldon Technologies" \
-      version="1.19.0-dev" \
+      version="1.20.0-dev" \
       release="1" \
       summary="Storage Initializer for Seldon Core" \
       description="Allows Seldon Core to download artifacts from cloud and local storage to a local volume"

diff --git a/components/rclone-storage-initializer/readme.md b/components/rclone-storage-initializer/readme.md
@@ -27,7 +27,7 @@ the `seldon-rclone-secret` using one of the configurations bellow and use follow
 
 ```yaml
 storageInitializer:
-  image: seldonio/rclone-storage-initializer:1.19.0-dev
+  image: seldonio/rclone-storage-initializer:1.20.0-dev
 
 predictiveUnit:
   defaultEnvSecretRefName: seldon-rclone-secret

diff --git a/docs-gb/README.md b/docs-gb/README.md
@@ -64,7 +64,7 @@ We provide optimized model servers for some of the most popular Deep Learning an
 You only have to upload your model binaries into your preferred object store, in this case we have a trained scikit-learn iris model in a Google bucket:
 
 ```console
-gs://seldon-models/v1.19.0-dev/sklearn/iris/model.joblib
+gs://seldon-models/v1.19.0/sklearn/iris/model.joblib
 ```
 
 Create a namespace to run your model in:
@@ -87,7 +87,7 @@ spec:
   predictors:
   - graph:
       implementation: SKLEARN_SERVER
-      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+      modelUri: gs://seldon-models/v1.19.0/sklearn/iris
       name: classifier
     name: default
     replicas: 1

diff --git a/docs-gb/deployments/abtests.md b/docs-gb/deployments/abtests.md
@@ -43,7 +43,7 @@ spec:
     traffic: 100
     graph:
       name: classifier
-      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
       implementation: SKLEARN_SERVER
   - name: candidate
     traffic: 0
@@ -357,7 +357,7 @@ spec:
   - name: default
     graph:
       name: classifier
-      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
       implementation: SKLEARN_SERVER
 ```
 

diff --git a/docs-gb/deployments/protocols.md b/docs-gb/deployments/protocols.md
@@ -40,7 +40,7 @@ spec:
   - graph:
       children: []
       implementation: SKLEARN_SERVER
-      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
       name: classifier
       parameters:
         - name: method

diff --git a/docs-gb/notebooks/argo_workflows_hdfs_batch.md b/docs-gb/notebooks/argo_workflows_hdfs_batch.md
@@ -150,7 +150,7 @@ spec:
   - graph:
       children: []
       implementation: SKLEARN_SERVER
-      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
       name: classifier
       logger:
         mode: all

diff --git a/docs-gb/notebooks/iris_explainer_poetry.md b/docs-gb/notebooks/iris_explainer_poetry.md
@@ -86,7 +86,7 @@ spec:
     graph:
       name: classifier
       implementation: SKLEARN_SERVER
-      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
 ```
 
     Overwriting iris.yaml
@@ -440,7 +440,7 @@ spec:
     graph:
       name: classifier
       implementation: SKLEARN_SERVER
-      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
     explainer:
       type: AnchorTabular
       modelUri: s3:explainers/iris/

diff --git a/docs-gb/notebooks/iter8-separate.md b/docs-gb/notebooks/iter8-separate.md
@@ -45,7 +45,7 @@ We will run a progressive rollout allowing Iter8 to control the traffic to the t
       - name: default
         graph:
           name: classifier
-          modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+          modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
           implementation: SKLEARN_SERVER
 
 

diff --git a/docs-gb/notebooks/iter8-single.md b/docs-gb/notebooks/iter8-single.md
@@ -47,7 +47,7 @@ We will run a progressive rollout allowing Iter8 to control the traffic to the t
         traffic: 100    
         graph:
           name: classifier
-          modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+          modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
           implementation: SKLEARN_SERVER
       - name: candidate
         traffic: 0

diff --git a/docs-gb/notebooks/knative_eventing_streaming.md b/docs-gb/notebooks/knative_eventing_streaming.md
@@ -29,7 +29,7 @@ spec:
   predictors:
   - graph:
       implementation: SKLEARN_SERVER
-      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
       name: simple-iris-model
       children: []
     name: default

diff --git a/docs-gb/notebooks/kubeflow_pipelines_batch.md b/docs-gb/notebooks/kubeflow_pipelines_batch.md
@@ -126,7 +126,7 @@ from kubernetes import client as k8s
 def nlp_pipeline(
         namespace="kubeflow",
         seldon_server="SKLEARN_SERVER",
-        model_path="gs://seldon-models/v1.19.0-dev/sklearn/iris",
+        model_path="gs://seldon-models/v1.20.0-dev/sklearn/iris",
         gateway_endpoint="istio-ingressgateway.istio-system.svc.cluster.local",
         retries=3,
         replicas=10,

diff --git a/docs-gb/notebooks/vegeta_bench_argo_workflows.md b/docs-gb/notebooks/vegeta_bench_argo_workflows.md
@@ -45,7 +45,7 @@ We will run a batch job that will set up a Seldon Deployment with 1 replicas and
     --set seldonDeployment.replicas="1" \
     --set seldonDeployment.serverWorkers="5" \
     --set seldonDeployment.serverThreads=1 \
-    --set seldonDeployment.modelUri="gs://seldon-models/v1.19.0-dev/sklearn/iris" \
+    --set seldonDeployment.modelUri="gs://seldon-models/v1.20.0-dev/sklearn/iris" \
     --set seldonDeployment.server="SKLEARN_SERVER" \
     --set seldonDeployment.apiType="rest|grpc" \
     --set seldonDeployment.requests.cpu="2000Mi" \

diff --git a/docs-gb/servers/overview.md b/docs-gb/servers/overview.md
@@ -23,7 +23,7 @@ spec:
       graph:
         name: classifier
         implementation: SKLEARN_SERVER
-        modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+        modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
 ```
 
 By default only public models published to Google Cloud Storage will be accessible. See below notes on how to configure credentials for AWS S3, Minio and other storage solutions.
@@ -156,7 +156,7 @@ spec:
     graph:
       name: classifier
       implementation: SKLEARN_SERVER
-      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
 ```
 
 The image name and other details will be added when this is deployed automatically.

diff --git a/docs-gb/servers/sklearn.md b/docs-gb/servers/sklearn.md
@@ -40,7 +40,7 @@ spec:
   - graph:
       children: []
       implementation: SKLEARN_SERVER
-      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
       name: classifier
     name: default
     replicas: 1
@@ -68,7 +68,7 @@ spec:
   - graph:
       children: []
       implementation: SKLEARN_SERVER
-      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
       name: classifier
       parameters:
         - name: method
@@ -104,7 +104,7 @@ spec:
   - graph:
       children: []
       implementation: SKLEARN_SERVER
-      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
       name: classifier
       parameters:
         - name: method
@@ -152,7 +152,7 @@ spec:
     graph:
       children: []
       implementation: SKLEARN_SERVER
-      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
       name: classifier
     name: default
     replicas: 1

diff --git a/docs-gb/upgrading.md b/docs-gb/upgrading.md
@@ -72,7 +72,7 @@ Following table provides a summary of Seldon Core / Kubernetes version compatibi
 
 It is always recommended to first upgrade Seldon Core to the latest supported version on your Kubernetes cluster and then upgrade the Kubernetes cluster.
 
-For example, if you using Seldon Core 1.15 on Kubernetes 1.24, it is recommended to **first upgrade Seldon Core to version 1.16** before upgrading your cluster to Kubernetes 1.25 and further.
+For example, if you're using Seldon Core 1.15 on Kubernetes 1.24, it is recommended to **first upgrade Seldon Core to version 1.16** before upgrading your cluster to Kubernetes 1.25 and further.
 
 ## Upgrading to 1.15
 
@@ -149,7 +149,7 @@ In this release we change status of [Seldon Core Analytics](../charts/seldon-cor
 
 ### Seldon Inference Payload Logging Changes
 
-The seldon executor logs the prediction request and respose payload to a configured URL post during the inference at each stage of a inference graph. These request/response pairs are now matched correctly to reflect the component's input/output at each node level as compared to pairing based on the tree structure of the inference graph previously. See [relevant issue](https://github.com/SeldonIO/seldon-core/issues/3873) for more details.
+The seldon executor logs the prediction request and response payload to a configured URL post during the inference at each stage of a inference graph. These request/response pairs are now matched correctly to reflect the component's input/output at each node level as compared to pairing based on the tree structure of the inference graph previously. See [relevant issue](https://github.com/SeldonIO/seldon-core/issues/3873) for more details.
 
 ### Alibi Detect Server Event Update
 
@@ -193,7 +193,7 @@ It is also possible to update these values on a per SeldonDeployment basis with
 
 ### Seldon Core Wrapper
 
- * With introduction of multi-processing in gRPC module the `SO_REUSEPORT` socket option is required. On certain Python distributions you may see `AttributeError: module 'socket' has no attribute 'SO_REUSEPORT'` error which would render gRPC endpoint non-operational. For Anaconda Python distributions we confirmed that upgrading to Python 3.7.10 or 3.8.10 removes the problem.
+ * With introduction of multiprocessing in gRPC module the `SO_REUSEPORT` socket option is required. On certain Python distributions you may see `AttributeError: module 'socket' has no attribute 'SO_REUSEPORT'` error which would render gRPC endpoint non-operational. For Anaconda Python distributions we confirmed that upgrading to Python 3.7.10 or 3.8.10 removes the problem.
 
 ### Server Updates
 
@@ -265,7 +265,7 @@ You can use and extend the [backwards compatibility notebook](../examples/backwa
 
 If you have a Kubernetes cluster with Seldon Core installed, and you want to upgrade the Kubernetes cluster, you have to carry out a set of manual steps due to the more strict validation that this version of Kubernetes introduced.
 
-To be more specific, we had to provide two versions of the CRD as part of the seldon core install helm chart. Similarly the CRD for Seldon Core in Kubernetes post-1.18 is actually differnt - namely, you can actually see that in version 1.3.0 we introduced new CRD changes in the helm chart via an IF statement to use a different CRD depending on the k8s version. Due to this, the path to upgrade from pre-1.18 to post-1.18 requires the following manual steps to be carried out:
+To be more specific, we had to provide two versions of the CRD as part of the seldon core installation helm chart. Similarly the CRD for Seldon Core in Kubernetes post-1.18 is actually differnt - namely, you can actually see that in version 1.3.0 we introduced new CRD changes in the helm chart via an IF statement to use a different CRD depending on the k8s version. Due to this, the path to upgrade from pre-1.18 to post-1.18 requires the following manual steps to be carried out:
 
 1. Start with kubernetes cluster pre 1.18 with seldon core pre-1.3.0
 2. Upgrade Kubernetes cluster to post 1.18 (seldon core CRD is now "invalid" but still installed as still in etcd)

diff --git a/examples/batch/argo-workflows-batch/helm-charts/seldon-batch-workflow/values.yaml b/examples/batch/argo-workflows-batch/helm-charts/seldon-batch-workflow/values.yaml
@@ -19,7 +19,7 @@ seldonDeployment:
   # Prepackaged model server to use [see https://docs.seldon.io/projects/seldon-core/en/latest/servers/overview.html]
   server: SKLEARN_SERVER
   # The URL for the model that is to be used
-  modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+  modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
   # The number of seldon deployment replicas to launch
   replicas: 2
   # Waiting time before checks for deployment to ensure kubernetes cluster registers create

diff --git a/examples/batch/hdfs-argo-workflows/deployment.yaml b/examples/batch/hdfs-argo-workflows/deployment.yaml
@@ -9,7 +9,7 @@ spec:
   - graph:
       children: []
       implementation: SKLEARN_SERVER
-      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
       name: classifier
       logger:
         mode: all

diff --git a/examples/batch/hdfs-argo-workflows/hdfs-batch.ipynb b/examples/batch/hdfs-argo-workflows/hdfs-batch.ipynb
@@ -239,7 +239,7 @@
     "  - graph:\n",
     "      children: []\n",
     "      implementation: SKLEARN_SERVER\n",
-    "      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris\n",
+    "      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris\n",
     "      name: classifier\n",
     "      logger:\n",
     "        mode: all\n",

diff --git a/examples/batch/kubeflow-pipelines-batch/README.ipynb b/examples/batch/kubeflow-pipelines-batch/README.ipynb
@@ -224,7 +224,7 @@
     "def nlp_pipeline(\n",
     "        namespace=\"kubeflow\",\n",
     "        seldon_server=\"SKLEARN_SERVER\",\n",
-    "        model_path=\"gs://seldon-models/v1.19.0-dev/sklearn/iris\",\n",
+    "        model_path=\"gs://seldon-models/v1.20.0-dev/sklearn/iris\",\n",
     "        gateway_endpoint=\"istio-ingressgateway.istio-system.svc.cluster.local\",\n",
     "        retries=3,\n",
     "        replicas=10,\n",

diff --git a/examples/batch/kubeflow-pipelines-batch/assets/seldon-batch-pipeline.py b/examples/batch/kubeflow-pipelines-batch/assets/seldon-batch-pipeline.py
@@ -9,7 +9,7 @@
 def nlp_pipeline(
     namespace="kubeflow",
     seldon_server="SKLEARN_SERVER",
-    model_path="gs://seldon-models/v1.19.0-dev/sklearn/iris",
+    model_path="gs://seldon-models/v1.20.0-dev/sklearn/iris",
     gateway_endpoint="istio-ingressgateway.istio-system.svc.cluster.local",
     retries=3,
     replicas=10,

diff --git a/examples/explainers/iris-explainer-poetry/iris-explainer.ipynb b/examples/explainers/iris-explainer-poetry/iris-explainer.ipynb
@@ -149,7 +149,7 @@
         "    graph:\n",
         "      name: classifier\n",
         "      implementation: SKLEARN_SERVER\n",
-        "      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris"
+        "      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris"
       ]
     },
     {
@@ -622,7 +622,7 @@
         "    graph:\n",
         "      name: classifier\n",
         "      implementation: SKLEARN_SERVER\n",
-        "      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris\n",
+        "      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris\n",
         "    explainer:\n",
         "      type: AnchorTabular\n",
         "      modelUri: s3:explainers/iris/\n",

diff --git a/examples/explainers/iris-explainer-poetry/iris-with-explainer.yaml b/examples/explainers/iris-explainer-poetry/iris-with-explainer.yaml
@@ -9,7 +9,7 @@ spec:
     graph:
       name: classifier
       implementation: SKLEARN_SERVER
-      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris
     explainer:
       type: AnchorTabular
       modelUri: s3:explainers/iris/

diff --git a/examples/explainers/iris-explainer-poetry/iris.yaml b/examples/explainers/iris-explainer-poetry/iris.yaml
@@ -9,4 +9,4 @@ spec:
     graph:
       name: classifier
       implementation: SKLEARN_SERVER
-      modelUri: gs://seldon-models/v1.19.0-dev/sklearn/iris
+      modelUri: gs://seldon-models/v1.20.0-dev/sklearn/iris