From e48554249a8db97a679bdcdc96c9607bdc7e01f1 Mon Sep 17 00:00:00 2001
From: Sivanantham <90966311+sivanantha321@users.noreply.github.com>
Date: Sun, 26 Nov 2023 20:45:21 +0530
Subject: [PATCH] Document servingruntime constraint introduced by
 kserve/kserve#3181 (#320)

* Document serving runtime constraint introduced by kserve/kserve#3181

Signed-off-by: Sivanantham Chinnaiyan <sivanantham.chinnaiyan@ideas2it.com>

* Set content type for predict/explainer curl requests

Signed-off-by: Sivanantham Chinnaiyan <sivanantham.chinnaiyan@ideas2it.com>

* Update docs/modelserving/servingruntimes.md

Signed-off-by: Dan Sun <dsun20@bloomberg.net>

---------

Signed-off-by: Sivanantham Chinnaiyan <sivanantham.chinnaiyan@ideas2it.com>
Signed-off-by: Dan Sun <dsun20@bloomberg.net>
Co-authored-by: Dan Sun <dsun20@bloomberg.net>
---
 .../serverless/kourier_networking/README.md   |  2 +-
 .../explainer/alibi/income/README.md          |  4 ++--
 .../explainer/alibi/moviesentiment/README.md  |  8 +++----
 .../inference_graph/image_pipeline/README.md  |  2 +-
 docs/modelserving/logger/logger.md            |  4 ++--
 docs/modelserving/servingruntimes.md          | 21 +++++++++++++++++++
 docs/modelserving/storage/azure/azure.md      |  2 +-
 docs/modelserving/storage/pvc/pvc.md          |  2 +-
 docs/modelserving/storage/s3/s3.md            |  2 +-
 docs/modelserving/storage/uri/uri.md          |  4 ++--
 docs/modelserving/v1beta1/amd/README.md       |  2 +-
 .../v1beta1/custom/custom_model/README.md     |  4 ++--
 docs/modelserving/v1beta1/lightgbm/README.md  |  2 +-
 docs/modelserving/v1beta1/pmml/README.md      |  2 +-
 .../v1beta1/rollout/canary-example.md         |  4 ++--
 docs/modelserving/v1beta1/spark/README.md     |  2 +-
 .../modelserving/v1beta1/tensorflow/README.md |  2 +-
 .../modelserving/v1beta1/torchserve/README.md | 12 +++++------
 .../v1beta1/transformer/feast/README.md       |  2 +-
 .../torchserve_image_transformer/README.md    |  4 ++--
 .../v1beta1/triton/bert/README.md             |  2 +-
 .../v1beta1/triton/torchscript/README.md      |  4 ++--
 22 files changed, 57 insertions(+), 36 deletions(-)

diff --git a/docs/admin/serverless/kourier_networking/README.md b/docs/admin/serverless/kourier_networking/README.md
index a6fea06f7..ea3a7cd14 100644
--- a/docs/admin/serverless/kourier_networking/README.md
+++ b/docs/admin/serverless/kourier_networking/README.md
@@ -130,7 +130,7 @@ Send a prediction request to the InferenceService and check the output.
 MODEL_NAME=pmml-demo
 INPUT_PATH=@./pmml-input.json
 SERVICE_HOSTNAME=$(kubectl get inferenceservice pmml-demo -o jsonpath='{.status.url}' | cut -d "/" -f 3)
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
 ```
 
 !!! success "Expected Output"
diff --git a/docs/modelserving/explainer/alibi/income/README.md b/docs/modelserving/explainer/alibi/income/README.md
index fcde7c209..61ab4c48c 100644
--- a/docs/modelserving/explainer/alibi/income/README.md
+++ b/docs/modelserving/explainer/alibi/income/README.md
@@ -57,7 +57,7 @@ SERVICE_HOSTNAME=$(kubectl get inferenceservice income -o jsonpath='{.status.url
 Test the predictor:
 
 ```bash
-curl -H "Host: $SERVICE_HOSTNAME" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d '{"instances":[[39, 7, 1, 1, 1, 1, 4, 1, 2174, 0, 40, 9]]}'
+curl -H "Host: $SERVICE_HOSTNAME" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d '{"instances":[[39, 7, 1, 1, 1, 1, 4, 1, 2174, 0, 40, 9]]}'
 ```
 
 You should receive the response showing the prediction is for low salary:
@@ -71,7 +71,7 @@ You should receive the response showing the prediction is for low salary:
 Now lets get an explanation for this:
 
 ```bash
-curl -H "Host: $SERVICE_HOSTNAME" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:explain -d '{"instances":[[39, 7, 1, 1, 1, 1, 4, 1, 2174, 0, 40, 9]]}'
+curl -H "Host: $SERVICE_HOSTNAME" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:explain -d '{"instances":[[39, 7, 1, 1, 1, 1, 4, 1, 2174, 0, 40, 9]]}'
 ```
 
 The returned explanation will be like:
diff --git a/docs/modelserving/explainer/alibi/moviesentiment/README.md b/docs/modelserving/explainer/alibi/moviesentiment/README.md
index c9909902b..c4e6f3149 100644
--- a/docs/modelserving/explainer/alibi/moviesentiment/README.md
+++ b/docs/modelserving/explainer/alibi/moviesentiment/README.md
@@ -56,7 +56,7 @@ SERVICE_HOSTNAME=$(kubectl get inferenceservice moviesentiment -o jsonpath='{.st
 Test the predictor on an example sentence:
 
 ```bash
-curl -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d '{"instances":["a visually flashy but narratively opaque and emotionally vapid exercise ."]}'
+curl -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d '{"instances":["a visually flashy but narratively opaque and emotionally vapid exercise ."]}'
 ```
 
 You should receive the response showing negative sentiment:
@@ -69,7 +69,7 @@ You should receive the response showing negative sentiment:
 Test on another sentence:
 
 ```bash
-curl -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d '{"instances":["a touching , sophisticated film that almost seems like a documentary in the way it captures an italian immigrant family on the brink of major changes ."]}'
+curl -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d '{"instances":["a touching , sophisticated film that almost seems like a documentary in the way it captures an italian immigrant family on the brink of major changes ."]}'
 ```
 
 You should receive the response showing positive sentiment:
@@ -83,7 +83,7 @@ Now lets get an explanation for the first sentence:
 
 
 ```bash
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:explain -d '{"instances":["a visually flashy but narratively opaque and emotionally vapid exercise ."]}'
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:explain -d '{"instances":["a visually flashy but narratively opaque and emotionally vapid exercise ."]}'
 ```
 
 !!! success "Expected Output"
@@ -234,7 +234,7 @@ kubectl create -f moviesentiment2.yaml
 and then ask for an explanation:
 
 ```bash
-curl -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:explain -d '{"instances":["a visually flashy but narratively opaque and emotionally vapid exercise ."]}'
+curl -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:explain -d '{"instances":["a visually flashy but narratively opaque and emotionally vapid exercise ."]}'
 ```
 
 !!! success "Expected Output"
diff --git a/docs/modelserving/inference_graph/image_pipeline/README.md b/docs/modelserving/inference_graph/image_pipeline/README.md
index 67c6eedd1..0fe13d501 100644
--- a/docs/modelserving/inference_graph/image_pipeline/README.md
+++ b/docs/modelserving/inference_graph/image_pipeline/README.md
@@ -146,7 +146,7 @@ The first step is to [determine the ingress IP and ports](../../../get_started/f
 Now, you can test the inference graph by sending the [cat](cat.json) and [dog image data](dog.json).
 ```bash
 SERVICE_HOSTNAME=$(kubectl get inferencegraph dog-breed-pipeline -o jsonpath='{.status.url}' | cut -d "/" -f 3)
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT} -d @./cat.json
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT} -d @./cat.json
 ```
 !!! success "Expected Output"
     ```{ .json .no-copy }
diff --git a/docs/modelserving/logger/logger.md b/docs/modelserving/logger/logger.md
index e0fcec504..2161bc302 100644
--- a/docs/modelserving/logger/logger.md
+++ b/docs/modelserving/logger/logger.md
@@ -77,7 +77,7 @@ MODEL_NAME=sklearn-iris
 INPUT_PATH=@./input.json
 SERVICE_HOSTNAME=$(kubectl get inferenceservice sklearn-iris -o jsonpath='{.status.url}' | cut -d "/" -f 3)
 
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
 ```
 
 !!! success "Expected Output"
@@ -283,7 +283,7 @@ MODEL_NAME=sklearn-iris
 INPUT_PATH=@./input.json
 SERVICE_HOSTNAME=$(kubectl get inferenceservice sklearn-iris -o jsonpath='{.status.url}' | cut -d "/" -f 3)
 
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
 ```
 
 !!! success "Expected Output"
diff --git a/docs/modelserving/servingruntimes.md b/docs/modelserving/servingruntimes.md
index 5bdd19989..0fcbe2afb 100644
--- a/docs/modelserving/servingruntimes.md
+++ b/docs/modelserving/servingruntimes.md
@@ -230,6 +230,27 @@ will be used for model deployment.
 - The serving runtime with priority takes precedence over the serving runtime with priority not specified.
 - Two model formats with same name and same model version cannot have the same priority.
 - If more than one serving runtime supports the model format and none of them specified the priority then, there is no guarantee _which_ runtime will be selected.
+- If multiple versions of a modelFormat are supported by a serving runtime, then it should have the same priority.
+  For example, Below shown serving runtime supports two versions of sklearn. It should have the same priority.
+  ```yaml
+    apiVersion: serving.kserve.io/v1alpha1
+    kind: ClusterServingRuntime
+    metadata:
+      name: mlserver
+    spec:
+        protocolVersions:
+          - v2
+        supportedModelFormats:
+          - name: sklearn
+            version: "0"
+            autoSelect: true
+            priority: 2
+          - name: sklearn
+            version: "1"
+            autoSelect: true
+            priority: 2
+  ...
+  ```
 
 !!! warning
     If multiple runtimes list the same format and/or version as auto-selectable and the priority is not specified, the runtime is selected based on the `creationTimestamp` i.e. the most recently created runtime is selected. So there is no guarantee _which_ runtime will be selected.
diff --git a/docs/modelserving/storage/azure/azure.md b/docs/modelserving/storage/azure/azure.md
index 8c2d69861..6f25393ac 100644
--- a/docs/modelserving/storage/azure/azure.md
+++ b/docs/modelserving/storage/azure/azure.md
@@ -108,7 +108,7 @@ SERVICE_HOSTNAME=$(kubectl get inferenceservice sklearn-azure -o jsonpath='{.sta
 
 MODEL_NAME=sklearn-azure
 INPUT_PATH=@./input.json
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
 ```
 
 !!! success "Expected Output"
diff --git a/docs/modelserving/storage/pvc/pvc.md b/docs/modelserving/storage/pvc/pvc.md
index 36013bf83..f79ec151a 100644
--- a/docs/modelserving/storage/pvc/pvc.md
+++ b/docs/modelserving/storage/pvc/pvc.md
@@ -134,7 +134,7 @@ SERVICE_HOSTNAME=$(kubectl get inferenceservice sklearn-pvc -o jsonpath='{.statu
 
 MODEL_NAME=sklearn-pvc
 INPUT_PATH=@./input.json
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
 ```
 
 !!! success "Expected Output"
diff --git a/docs/modelserving/storage/s3/s3.md b/docs/modelserving/storage/s3/s3.md
index fe454eca2..fe9fcb443 100644
--- a/docs/modelserving/storage/s3/s3.md
+++ b/docs/modelserving/storage/s3/s3.md
@@ -131,7 +131,7 @@ SERVICE_HOSTNAME=$(kubectl get inferenceservice mnist-s3 -o jsonpath='{.status.u
 
 MODEL_NAME=mnist-s3
 INPUT_PATH=@./input.json
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
 ```
 
 !!! success "Expected Output"
diff --git a/docs/modelserving/storage/uri/uri.md b/docs/modelserving/storage/uri/uri.md
index ff3f3232e..93d274f27 100644
--- a/docs/modelserving/storage/uri/uri.md
+++ b/docs/modelserving/storage/uri/uri.md
@@ -137,7 +137,7 @@ SERVICE_HOSTNAME=$(kubectl get inferenceservice sklearn-from-uri -o jsonpath='{.
 
 MODEL_NAME=sklearn-from-uri
 INPUT_PATH=@./input.json
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
 ```
 
 !!! success "Expected Output"
@@ -265,7 +265,7 @@ SERVICE_HOSTNAME=$(kubectl get inferenceservice tensorflow-from-uri-gzip -o json
 
 MODEL_NAME=tensorflow-from-uri-gzip
 INPUT_PATH=@./input.json
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
 ```
 
 !!! success "Expected Output"
diff --git a/docs/modelserving/v1beta1/amd/README.md b/docs/modelserving/v1beta1/amd/README.md
index dbcaaf521..c5e05ce5f 100644
--- a/docs/modelserving/v1beta1/amd/README.md
+++ b/docs/modelserving/v1beta1/amd/README.md
@@ -100,7 +100,7 @@ Assuming that `INGRESS_HOST`, `INGRESS_PORT`, and `SERVICE_HOSTNAME` have been d
 ```bash
 export MODEL_NAME=mnist
 export INPUT_DATA=@./input.json
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v2/models/${MODEL_NAME}/infer -d ${INPUT_DATA}
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v2/models/${MODEL_NAME}/infer -d ${INPUT_DATA}
 ```
 
 This shows the response from the server in KServe's v2 API format.
diff --git a/docs/modelserving/v1beta1/custom/custom_model/README.md b/docs/modelserving/v1beta1/custom/custom_model/README.md
index 62fb26e3d..23a6af1a0 100644
--- a/docs/modelserving/v1beta1/custom/custom_model/README.md
+++ b/docs/modelserving/v1beta1/custom/custom_model/README.md
@@ -81,7 +81,7 @@ docker run -ePORT=8080 -p8080:8080 ${DOCKER_USER}/custom-model:v1
 
 Send a test inference request locally with [input.json](./input.json)
 ```bash
-curl localhost:8080/v1/models/custom-model:predict -d @./input.json
+curl -H "Content-Type: application/json" localhost:8080/v1/models/custom-model:predict -d @./input.json
 ```
 !!! success "Expected Output"
     ```{ .json .no-copy }
@@ -146,7 +146,7 @@ MODEL_NAME=custom-model
 INPUT_PATH=@./input.json
 SERVICE_HOSTNAME=$(kubectl get inferenceservice ${MODEL_NAME} -o jsonpath='{.status.url}' | cut -d "/" -f 3)
 
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/${MODEL_NAME}:predict -d $INPUT_PATH
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/${MODEL_NAME}:predict -d $INPUT_PATH
 ```
 
 !!! success "Expected Output"
diff --git a/docs/modelserving/v1beta1/lightgbm/README.md b/docs/modelserving/v1beta1/lightgbm/README.md
index 513bbaf4d..d73d34e7c 100644
--- a/docs/modelserving/v1beta1/lightgbm/README.md
+++ b/docs/modelserving/v1beta1/lightgbm/README.md
@@ -101,7 +101,7 @@ To test the deployed model the first step is to [determine the ingress IP and po
 MODEL_NAME=lightgbm-iris
 INPUT_PATH=@./iris-input.json
 SERVICE_HOSTNAME=$(kubectl get inferenceservice lightgbm-iris -o jsonpath='{.status.url}' | cut -d "/" -f 3)
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
 ```
 
 !!! success "Expected Output"
diff --git a/docs/modelserving/v1beta1/pmml/README.md b/docs/modelserving/v1beta1/pmml/README.md
index 5a105afab..b83e7698a 100644
--- a/docs/modelserving/v1beta1/pmml/README.md
+++ b/docs/modelserving/v1beta1/pmml/README.md
@@ -63,7 +63,7 @@ You can see an example payload below. Create a file named `iris-input.json` with
 MODEL_NAME=pmml-demo
 INPUT_PATH=@./iris-input.json
 SERVICE_HOSTNAME=$(kubectl get inferenceservice pmml-demo -o jsonpath='{.status.url}' | cut -d "/" -f 3)
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
 ```
 
 !!! success "Expected Output"
diff --git a/docs/modelserving/v1beta1/rollout/canary-example.md b/docs/modelserving/v1beta1/rollout/canary-example.md
index d6fe1cdd9..0c7ad956d 100644
--- a/docs/modelserving/v1beta1/rollout/canary-example.md
+++ b/docs/modelserving/v1beta1/rollout/canary-example.md
@@ -249,13 +249,13 @@ MODEL_NAME=sklearn-iris
 curl the latest revision
 
 ```bash
-curl -v -H "Host: latest-${MODEL_NAME}-predictor-default.kserve-test.example.com" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d @./iris-input.json
+curl -v -H "Host: latest-${MODEL_NAME}-predictor-default.kserve-test.example.com" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d @./iris-input.json
 
 ```
 
 or curl the previous revision
 
 ```bash
-curl -v -H "Host: prev-${MODEL_NAME}-predictor-default.kserve-test.example.com" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d @./iris-input.json
+curl -v -H "Host: prev-${MODEL_NAME}-predictor-default.kserve-test.example.com" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d @./iris-input.json
 
 ```
diff --git a/docs/modelserving/v1beta1/spark/README.md b/docs/modelserving/v1beta1/spark/README.md
index ae30969f2..8e3f4fc38 100644
--- a/docs/modelserving/v1beta1/spark/README.md
+++ b/docs/modelserving/v1beta1/spark/README.md
@@ -109,7 +109,7 @@ You can see an example payload below. Create a file named `iris-input.json` with
 MODEL_NAME=spark-pmml
 INPUT_PATH=@./iris-input.json
 SERVICE_HOSTNAME=$(kubectl get inferenceservice spark-pmml -o jsonpath='{.status.url}' | cut -d "/" -f 3)
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
 ```
 
 !!! success "Expected Output"
diff --git a/docs/modelserving/v1beta1/tensorflow/README.md b/docs/modelserving/v1beta1/tensorflow/README.md
index 631b6b290..68a7b45b1 100644
--- a/docs/modelserving/v1beta1/tensorflow/README.md
+++ b/docs/modelserving/v1beta1/tensorflow/README.md
@@ -60,7 +60,7 @@ file can be downloaded [here](./input.json).
 MODEL_NAME=flower-sample
 INPUT_PATH=@./input.json
 SERVICE_HOSTNAME=$(kubectl get inferenceservice ${MODEL_NAME} -o jsonpath='{.status.url}' | cut -d "/" -f 3)
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict -d $INPUT_PATH
 ```
 
 !!! success "Expected Output"
diff --git a/docs/modelserving/v1beta1/torchserve/README.md b/docs/modelserving/v1beta1/torchserve/README.md
index 3217931c6..e9fde7458 100644
--- a/docs/modelserving/v1beta1/torchserve/README.md
+++ b/docs/modelserving/v1beta1/torchserve/README.md
@@ -158,7 +158,7 @@ SERVICE_HOSTNAME=$(kubectl get inferenceservice torchserve -o jsonpath='{.status
 You can use [image converter](https://github.com/kserve/kserve/tree/master/docs/samples/v1beta1/torchserve/v1/imgconv) to convert the images to base64 byte array, for other models please refer to [input request](https://github.com/pytorch/serve/tree/master/kubernetes/kserve/kf_request_json).
 
 ```bash
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/${MODEL_NAME}:predict -d @./mnist.json
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/${MODEL_NAME}:predict -d @./mnist.json
 ```
 
 !!! success "Expected Output"
@@ -194,7 +194,7 @@ curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1
 To get model explanation:
 
 ```bash
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/mnist:explain -d @./mnist.json
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/mnist:explain -d @./mnist.json
 ```
 
 !!! success "Expected Output"
@@ -261,7 +261,7 @@ SERVICE_HOSTNAME=$(kubectl get inferenceservice torchserve-mnist-v2 -o jsonpath=
 You can send both **byte array** and **tensor** with v2 protocol, for byte array use [image converter](https://github.com/kserve/kserve/tree/master/docs/samples/v1beta1/torchserve/v2/bytes_conv) to convert the image to byte array input. Here we use the [mnist_v2_bytes.json](./mnist_v2_bytes.json) file to run an example inference.
 
 ```bash
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v2/models/${MODEL_NAME}/infer -d @./mnist_v2_bytes.json
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v2/models/${MODEL_NAME}/infer -d @./mnist_v2_bytes.json
 ```
 
 !!! success "Expected Output"
@@ -273,7 +273,7 @@ For tensor input use the [tensor image converter](https://github.com/kserve/kser
 tensor input and here we use the [mnist_v2.json](./mnist_v2.json) file to run an example inference.
 
 ```bash
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v2/models/${MODEL_NAME}/infer -d @./mnist_v2.json
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v2/models/${MODEL_NAME}/infer -d @./mnist_v2.json
 ```
 
 !!! success "Expected Output"
@@ -287,7 +287,7 @@ To get the model explanation with v2 explain endpoint:
 
 ```bash
 MODEL_NAME=mnist
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v2/models/mnist/explain -d @./mnist_v2.json
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v2/models/mnist/explain -d @./mnist_v2.json
 ```
 
 !!! success "Expected Output"
@@ -564,7 +564,7 @@ kubectl patch isvc torchserve --type='json' -p '[{"op": "replace", "path": "/spe
 ```
 
 ```bash
-curl -v -H "Host: latest-torchserve-predictor-default.default.example.com" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/${MODEL_NAME}:predict -d @./mnist.json
+curl -v -H "Host: latest-torchserve-predictor-default.default.example.com" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/${MODEL_NAME}:predict -d @./mnist.json
 ```
 
 !!! success "Expected Output"
diff --git a/docs/modelserving/v1beta1/transformer/feast/README.md b/docs/modelserving/v1beta1/transformer/feast/README.md
index 80310c4d0..f4a405b19 100644
--- a/docs/modelserving/v1beta1/transformer/feast/README.md
+++ b/docs/modelserving/v1beta1/transformer/feast/README.md
@@ -275,7 +275,7 @@ MODEL_NAME=sklearn-driver-transformer
 INPUT_PATH=@./driver-input.json
 SERVICE_HOSTNAME=$(kubectl get inferenceservice $SERVICE_NAME -o jsonpath='{.status.url}' | cut -d "/" -f 3)
 
-curl -v -H "Host: ${SERVICE_HOSTNAME}" -d $INPUT_PATH http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" -d $INPUT_PATH http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict
 ```
 
 !!! success "Expected output"
diff --git a/docs/modelserving/v1beta1/transformer/torchserve_image_transformer/README.md b/docs/modelserving/v1beta1/transformer/torchserve_image_transformer/README.md
index f199046a6..f20291905 100644
--- a/docs/modelserving/v1beta1/transformer/torchserve_image_transformer/README.md
+++ b/docs/modelserving/v1beta1/transformer/torchserve_image_transformer/README.md
@@ -182,7 +182,7 @@ MODEL_NAME=mnist
 INPUT_PATH=@./input.json
 SERVICE_HOSTNAME=$(kubectl get inferenceservice $SERVICE_NAME -o jsonpath='{.status.url}' | cut -d "/" -f 3)
 
-curl -v -H "Host: ${SERVICE_HOSTNAME}" -d $INPUT_PATH http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" -d $INPUT_PATH http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict
 ```
 
 !!! success "Expected Output"
@@ -306,7 +306,7 @@ MODEL_NAME=cifar10
 INPUT_PATH=@./image.json
 SERVICE_HOSTNAME=$(kubectl get inferenceservice $SERVICE_NAME -o jsonpath='{.status.url}' | cut -d "/" -f 3)
 
-curl -v -H "Host: ${SERVICE_HOSTNAME}" -d $INPUT_PATH http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" -d $INPUT_PATH http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict
 ```
 
 !!! success "Expected Output"
diff --git a/docs/modelserving/v1beta1/triton/bert/README.md b/docs/modelserving/v1beta1/triton/bert/README.md
index 01a990ac5..b6f656fa8 100644
--- a/docs/modelserving/v1beta1/triton/bert/README.md
+++ b/docs/modelserving/v1beta1/triton/bert/README.md
@@ -170,7 +170,7 @@ MODEL_NAME=bert-v2
 INPUT_PATH=@./input.json
 SERVICE_HOSTNAME=$(kubectl get inferenceservices bert-v2 -o jsonpath='{.status.url}' | cut -d "/" -f 3)
 
-curl -v -H "Host: ${SERVICE_HOSTNAME}" -d $INPUT_PATH http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" -d $INPUT_PATH http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/$MODEL_NAME:predict
 ```
 
 !!! success "Expected output"
diff --git a/docs/modelserving/v1beta1/triton/torchscript/README.md b/docs/modelserving/v1beta1/triton/torchscript/README.md
index e1579142a..11df2e971 100644
--- a/docs/modelserving/v1beta1/triton/torchscript/README.md
+++ b/docs/modelserving/v1beta1/triton/torchscript/README.md
@@ -163,7 +163,7 @@ curl -O https://raw.githubusercontent.com/kserve/kserve/master/docs/samples/v1be
 MODEL_NAME=cifar10
 INPUT_PATH=@./input.json
 SERVICE_HOSTNAME=$(kubectl get inferenceservice torchscript-cifar10 -o jsonpath='{.status.url}' | cut -d "/" -f 3)
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v2/models/${MODEL_NAME}/infer -d $INPUT_PATH
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v2/models/${MODEL_NAME}/infer -d $INPUT_PATH
 ```
 !!! success "Expected Output"
     ```{ .bash .no-copy }
@@ -463,7 +463,7 @@ INPUT_PATH=@./image.json
 
 SERVICE_HOSTNAME=$(kubectl get inferenceservice $SERVICE_NAME -o jsonpath='{.status.url}' | cut -d "/" -f 3)
 
-curl -v -H "Host: ${SERVICE_HOSTNAME}" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/${MODEL_NAME}:predict -d $INPUT_PATH
+curl -v -H "Host: ${SERVICE_HOSTNAME}" -H "Content-Type: application/json" http://${INGRESS_HOST}:${INGRESS_PORT}/v1/models/${MODEL_NAME}:predict -d $INPUT_PATH
 ```
 
 !!! success "Expected Output"