context stringlengths 24 500 | question stringlengths 36 186 | response stringlengths 44 2.17k |
|---|---|---|
d e v r e s e r s t h g i r
l l
A
. y n a p m o c e t a
i l
ffi a P A S n a r o E S P A S 5 2 0 2 ©
INTERNAL
2025-02-25
Generative AI Hub with SAP AI Core
THE BEST RUN | What is the date of the recent update (2025-02-25) related to the Generative AI Hub with SAP AI Core? | The date of the recent update is 2025-02-25. |
Content
1
Generative AI Hub in SAP AI Core. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3
1.1
Quick Start. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Get an Auth Token. . . . . . ... | Is the Generative AI Hub in SAP AI Core a useful tool for developers? | The document does not provide information about the usefulness of the Generative AI Hub in SAP AI Core for developers. It only mentions the hub's existence and provides a brief outline of its content, including a "Quick Start" section.
However, based on general knowledge, a Generative AI Hub in a platform like SAP AI... |
Create a Deployment for Orchestration in SAP AI Core. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Consume Models with the Harmonized API. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Next Steps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... | What are the steps involved in creating a Deployment for Orchestration in SAP AI Core? | Unfortunately, the provided document snippet does not explicitly mention the steps involved in creating a Deployment for Orchestration in SAP AI Core. However, based on the title "Create a Deployment for Orchestration in SAP AI Core" and the chapter number "10", it can be inferred that the chapter 10 of the document li... |
Accessing Generative AI Models Through Global Scenarios. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Orchestration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Foundation Models. . . . . . . . . . . . . . . . . . . . . . .... | What are the key concepts discussed in the chapter on Accessing Generative AI Models Through Global Scenarios? | Unfortunately, the provided snippet does not contain information about the chapter on Accessing Generative AI Models Through Global Scenarios. The snippet only mentions the chapter title and its page number (12), but does not provide any details about the content of the chapter.
However, based on the title, it can be ... |
Supported Models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .120
1.4
Additional Concepts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .121
Prompt Registry. . . . . . . . . . . .... | What models are supported in this context? | Unfortunately, the provided document does not mention "Bart models." However, it does mention "Supported Models" on page 10, but the content of that page is not provided. |
Service Plans. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .131
Models and Scenarios in the Generative AI Hub. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
Metering and Pricing for the Generative AI Hub. . . . .... | What type of service plans are available in the Generative AI Hub? | Unfortunately, the provided snippet does not explicitly mention the type of service plans available in the Generative AI Hub. However, based on the context, it seems that the Generative AI Hub is related to a cloud-based service, possibly offering various plans or tiers for users to access its features and capabilities... |
Add a Service Plan for Generative AI to the Global Account. . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Update an Existing Plan of SAP AI Core to Include Generative AI. . . . . . . . . . . . . . . . . . . . . . . . 142
1.6
Resource Plans for Large Language Models. . . . . . . . . . . . . . . . . . . . ... | Should we add a Service Plan for Generative AI to the Global Account? | Yes, according to the document, the answer is affirmative. The document explicitly states: "Add a Service Plan for Generative AI to the Global Account............................................................................................................................................................................. |
1 Generative AI Hub in SAP AI Core
The generative AI hub incorporates generative AI into your AI activities in SAP AI Core and SAP AI Launchpad. | What are the key features and functionalities of the Generative AI Hub in SAP AI Core? | Based on the provided information, the key features and functionalities of the Generative AI Hub in SAP AI Core are:
1. Incorporation of generative capabilities into AI Core.
2. Allows users to incorporate generative capabilities into their AI projects in AI Core and AI Orchestrator.
3. Enables users to use generative... |
Generative AI models are self-supervised, deep learning models trained on vast amounts of unlabeled data. They use AI technology and industrial-scale computational resources to learn complex patterns and semantic knowledge bases. These models excel in tasks like natural language processing (NLP). By parsing inputs, suc... | What are the key characteristics of generative AI models that enable them to excel in various tasks? | The key characteristics of generative AI models that enable them to excel in various tasks are:
1. Self-supervised: They are trained on vast amounts of unlabeled data.
2. Deep learning models: They utilize complex neural networks to learn patterns.
3. Industrial-scale computational resources: They require significant ... |
formats and output modes. | What are the different formats and output modes used to display data? | However, I don't see a document provided. Please share the document, and I'll be happy to help you understand the different medications and output modes used to display data. |
Generative AI models are general by design, but you can fine-tune them with additional embeddings. This makes them suitable for specialized or domain-specific use cases.
SAP AI Core and the generative AI hub help you to integrate LLMs and AI into new business processes in a cost-efficient manner.
Generative AI Hub Ar... | Can generative AI models be fine-tuned for domain-specific tasks? | Yes, generative AI models can be fine-tuned for domain-specific tasks with additional embeddings. |
Prerequisites
The generative AI hub is available only with the extended and sap-internal service plans.
Generative AI Hub with SAP AI Core Generative AI Hub in SAP AI Core
INTERNAL
3
Related Information
Service Plans [page 131]
1.1.1 Get an Auth Token
Using Postman
Prerequisites
You have downloaded and instal... | What are the service plans available for accessing the Generative AI hub in SAP AI Core? | According to the document, the Generative AI hub is available only with the "extended" and "sap-internal" service plans. |
Procedure
1. Download the JSON collection from Orchestration JSON.
2.
In Postman, click Import, select the JSON file, and choose Import to start the import.
3. After the import is complete, expand the collection and navigate to Get Auth Token. | What is the first step to start the JSON collection import process? | The first step to start the JSON collection import process is to download the JSON collection from Orchestration JSON. |
4. Select the LLM Orchestration environment within Postman and configure it using the following values: • The LLM Orch Url is the URL of your orchestration deployment. For more information, see Create a Deployment for Orchestration in SAP AI Core [page 6].
The Client Id is the XSUAA client id of your AI API connection... | What are the steps to configure postman for LLM orchestration? | To configure Postman for LLM orchestration, follow these steps:
1. Select the LLM Orchestration environment within Postman.
2. Configure the environment using the following values:
- LLM Orch URL: The URL of your orchestration deployment.
- Client Id: The XSUAA client id of your AI API connection.
- Client Se... |
5. To retrieve an authentication token, send a GET request to Auth Token.
The response includes an access token, which is saved as the Token variable in the Postman environment. Check that the access token has been added correctly to the variable and if not, copy and paste it from the response into the variable. | Is the access token successfully saved in the Postman environment variable after sending a GET request to Auth Token for authentication? | Yes, the access token is successfully saved in the Postman environment variable after sending a GET request to Auth Token for authentication. The document states that the access token is saved as the Token variable in the Postman environment. |
Next Steps
Send a POST request using the Simple LLM Call from the Postman collection to try out the templating mechanism of orchestration.
4
INTERNAL
Generative AI Hub with SAP AI Core Generative AI Hub in SAP AI Core
Using Curl
Start by setting the required environment variables, which you can get from your SAP ... | What are the required environment variables needed to send a POST request using the Simple LLM Call from the Postman collection? | The required environment variables needed to send a POST request using the Simple LLM Call from the Postman collection can be obtained from your SAP AI Core instance. |
Prerequisites
curl is likely to be installed on your operating system by default. To check, open a command prompt and enter curl -V. If curl isn't installed, download and install it from https://curl.se/
.
Note
On macOS, you may need to install jq so that you can follow the curl commands.
1.
Install brew from h... | What is curl and how do I check if it's installed on my operating system? | According to the document, curl is likely to be installed on your operating system by default. To check if it's installed, you can open a command prompt and enter `curl -V`. If curl isn't installed, you can download and install it from https://curl.se/. |
Procedure
1. Set up your environment as follows:
AI_API_URL=<YOUR AI API URL> AUTH_URL=<YOUR AUTH URL> CLIENT_ID=<YOUR CLIENT ID> CLIENT_SECRET=<YOUR CLIENT SECRET> RESOURCE_GROUP=default
2. Obtain the auth token by sending the following request:
curl --request POST \ --url "${AUTH_URL}/oauth/token" \ --header 'con... | Can I obtain an auth token by sending a POST request to the specified AUTH_URL with the grant type as client_credentials and the client ID and client secret? | Yes, you can obtain an auth token by sending a POST request to the specified AUTH_URL with the grant type as client_credentials and the client ID and client secret. This is explicitly stated in the provided procedure. |
The response includes an access token.
{"access_token":"ey...", "expires_in":7199, "jti":"...", "token_type":"bearer"}
3. Set the token to use it in the following steps:
TOKEN=ey...
Generative AI Hub with SAP AI Core Generative AI Hub in SAP AI Core
INTERNAL
5
1.1.2 Create a Deployment for Orchestration in SAP A... | What information should I use as the token when setting the TOKEN variable in the subsequent steps? | You should use the entire access token, including the "ey..." prefix, as the token when setting the TOKEN variable.
TOKEN=ey... |
Prerequisites
You have an SAP AI Core service instance and service key. For more information, see SAP AI Core Initial Setup Documentation.
You’re using the sap-internal or extended service plan. For more information, see Service Plans [page 131], as well as Add a Service Plan for Generative AI to the Global Account [... | What are the prerequisites to get started with SAP AI Core? | To get started with SAP AI Core, the prerequisites are:
1. An SAP AI Core service instance and service key.
2. An SAP AI Core service plan of either sap-internal or extended. |
You have completed the client authorization for your preferred user interface. For more information, see Use a Service Key.
You have at least one orchestration-compatible deployment for a generative AI model running. For more information, see and Create a Deployment for Orchestration in SAP AI Core [page 6]. | Is your client authorization set up for the preferred user interface to ensure a smooth experience with the generative AI model? | Yes, the client authorization is set up for the preferred user interface, as indicated by the completion of the client authorization process. |
Context
You create a deployment to make orchestration capabilities available for use. After the deployment is complete, you get a deploymentUrl. You can use this URL across your organization to access orchestration in the generative AI hub.
For more information, see Consumption of GenAI Models Using Orchestration – A... | Can I use the deploymentUrl across my entire organization to access orchestration in the generative AI hub after the deployment is complete? | Yes, you can use the deploymentUrl across your entire organization to access orchestration in the generative AI hub after the deployment is complete. |
Procedure
1. Check that you have access to the orchestration scenario containing generative AI by sending a GET
request to {{apiurl}}/v2/lm/scenarios.
Set the Authorization header with Bearer $TOKEN and set your resource group. | What is the correct URL for accessing the orchestration scenario containing generative AI by sending a GET request to the API? | The correct URL for accessing the orchestration scenario containing generative AI by sending a GET request to the API is:
{{apiurl}}/v2/lm/scenarios |
Note
You must use the same resource group for all of your generative AI activities. To use a different resource group, these steps must be repeated for each resource group.
The scenarios listed contain a scenario with the id orchestration.
2. Create a configuration by sending a POST request to the endpoint {{apiur... | Will I need to use the same resource group for all of my generative AI activities? | Yes, you must use the same resource group for all of your generative AI activities. |
configurations.
6
INTERNAL
Generative AI Hub with SAP AI Core Generative AI Hub in SAP AI Core
Include the following parameters:
name is your free choice of identifier. • executableId must be orchestration. • scenarioId must be orchestration. • versionId is your own version reference. Sample Code | What are the required parameters to configure a Generative AI Hub in SAP AI Core? | The required parameters to configure a Generative AI Hub in SAP AI Core are:
1. name (a free choice of identifier)
2. executableId (must be "orchestration")
3. scenarioId (must be "orchestration")
4. versionId (your own version reference) |
{ "name": "yourNameChoice", "executableId": "orchestration", "scenarioId": "orchestration", "versionId": "0.0.1", }
You receive a unique configurationId in the response.
3. Create a deployment by sending a POST request to the endpoint {{apiurl}}/v2/lm/deployments.
Include the configurationId from the previous step i... | Can you provide the request body for the POST request to create a deployment including the unique configurationId received in the previous step? | The request body for the POST request to create a deployment is:
```json
{
"configurationId": "yourConfigurationId"
}
```
Replace `"yourConfigurationId"` with the actual unique `configurationId` received in the previous step. |
deployments.
Using curl | Can deployments be successfully executed using curl? | Unfortunately, you didn't provide a document. However, I can provide a general answer to your question.
Yes, deployments can be successfully executed using curl.
You can use curl to deploy applications to a Kubernetes cluster by sending a POST request to the Kubernetes API server. The request body should contain the... |
Procedure
1. Create a configure for the orchestration deployment.
Orchestration is exposed via the global scenario orchestration and the executable orchestration.
a. Send the following request:
curl --request POST "$AI_API_URL/v2/lm/configurations" \ --header "Authorization: Bearer $TOKEN" \ --header "ai-resource-g... | Is there a configuration for the orchestration deployment that can be created with a POST request to the AI API URL? | Yes, there is a configuration for the orchestration deployment that can be created with a POST request to the AI API URL.
The document provides a curl command that sends a POST request to the AI API URL to create a configuration for the orchestration deployment. The command includes the necessary headers and a JSON p... |
Generative AI Hub with SAP AI Core Generative AI Hub in SAP AI Core
INTERNAL
7
The response appears as follows:
Output Code
{ "id": "f7ac7f77-e70f-4e9c-86b5-1504b44fe789", "message": "Configuration created" }
b. Take the id of the configuration and set it as an environment variable.
ORCH_CONFIG_ID=f7ac7f77-e70... | What is the correct command to set the configuration id as an environment variable in the context of the Generative AI Hub in SAP AI Core? | The correct command to set the configuration id as an environment variable is:
```bash
ORCH_CONFIG_ID=f7ac7f77-e70f-4e9c-86b5-1504b44fe789
```
This is explicitly stated in the document as step 2b. |
curl --request POST $AI_API_URL/v2/lm/deployments \ --header 'content-type: application/json' \ --header "Authorization: Bearer $TOKEN" \ --header "ai-resource-group: $RESOURCE_GROUP" \ --data-raw "{ \"configurationId\": \"$ORCH_CONFIG_ID\" }"
The response appears as follows:
Output Code
{ "deploymentUrl": "", "id... | What is the value of the "id" key in the JSON response from the POST request to deploy a model? | The value of the "id" key in the JSON response from the POST request to deploy a model is "d4168482710c6cf9". |
ORCH_DEPLOYMENT_ID=d4168482710c6cf9
3. Wait for the deployment to start.
It takes a few minutes for the orchestration deployment to reach the status RUNNING. You can check the status of the deployment using the deployment id.
curl --request GET "$AI_API_URL/v2/lm/deployments/$ORCH_DEPLOYMENT_ID" \ --header "Authoriz... | What is the status of the deployment with ID d4168482710c6cf9? | The status of the deployment with ID d4168482710c6cf9 can be checked using the provided curl command. However, the document does not provide the actual status of the deployment.
To answer the question, we would need to run the curl command and retrieve the status of the deployment. Based on the information provided, ... |
If the output looks as follows (that is, the status is RUNNING), the deployment is ready. If not, try again after a couple of minutes.
{ "configurationId": "f7ac7f77-e70f-4e9c-86b5-1504b44fe789", "configurationName": "orchestration-config", "createdAt": "2024-06-20T14:40:16Z", "deploymentUrl": , "details": { "resource... | Is the deployment of the Generative AI Hub in SAP AI Core ready? | Based on the provided information, the status of the deployment is RUNNING. However, the document does not explicitly state whether the deployment is ready or not. It only mentions that if the output looks as follows (with the status RUNNING), the deployment is ready, implying that the deployment is not yet ready if th... |
"scaling": { "backend_details": {} } }, "id": "d4468482710c6cf9", "lastOperation": "CREATE", "latestRunningConfigurationId": "f7ac7f77-e70f-4e9c-86b5-1504b44fe789", "modifiedAt": "2024-06-20T14:48:03Z", "scenarioId": "llm-orchestration", "startTime": "2024-06-20T14:41:22Z", "status": "RUNNING", "submissionTime": "2024-... | What is the value of the deploymentUrl in the given JSON object? | The value of the deploymentUrl is not present in the given JSON object. The object contains information about a scenario, but it does not include a deploymentUrl. |
The deploymentUrl is available only when the deployment has the status RUNNING.
ORCH_DEPLOYMENT_URL="<deployment_url>"
1.1.3 Consume Models with the Harmonized API
In this section, we will provide a minimal inference call without any orchestration modules.
A minimal call to orchestration contains only configuration... | Does the deployment URL become available when the deployment is in the RUNNING status? | Yes, the deployment URL is available only when the deployment has the status RUNNING. |
curl --request POST $ORCH_DEPLOYMENT_URL/completion \ --header 'content-type: application/json' \ --header "Authorization: Bearer $TOKEN" \ --header "ai-resource-group: $RESOURCE_GROUP" \ --data-raw '{ "orchestration_config": { "module_configurations": { "templating_module_config": { "template": [ { "role": "user", "co... | Can we use an example of a POST request to demonstrate the curl command used to send a completion request to the Orchestration API? | Yes, we can use an example of a POST request to demonstrate the curl command used to send a completion request to the Orchestration API.
Here's an example of the curl command:
```bash
curl --request POST $ORCH_DEPLOYMENT_URL/completion \
--header 'content-type: application/json' \
--header "Authorization: Bearer... |
50, "temperature": 0.1, "frequency_penalty": 0, "presence_penalty": 0 }, "model_version": "latest" } } }, "input_params": { "text": "Orchestration is Working!", "language": "German" } | Is the given JSON response indicating that the model has successfully generated text in the German language? | Based on the provided JSON response, it appears that the model has successfully generated text in the German language. This is indicated by the "input_params" section, where the "language" is set to "German". However, there is no explicit indication that the model has generated text. The "output" section contains a "mo... |
Generative AI Hub with SAP AI Core Generative AI Hub in SAP AI Core
INTERNAL
9
}' | What does a Generative AI Hub in SAP AI Core look like? | Based on the provided information, it appears that the Generative AI Hub in SAP AI Core is not fully described in the snippet. However, I can provide a general understanding of what a Generative AI Hub might look like based on the context.
A Generative AI Hub in SAP AI Core likely serves as a centralized platform for ... |
This request configures the templating module with a single user message with two parameters: text and language. The language parameter is also configured with English as the default. The LLM module is configured to use gpt-4o-mini in the latest available version and a set of model parameters. The input_params field co... | What is the default value of the language parameter in this templating module configuration? | The default value of the language parameter is English. |
The response contains a request_id, the module results from each module that was executed, and the orchestration_result, which includes the response of the call to the model.
Output Code | Can we add a request_id to track requests, store module results for auditing purposes, and display the orchestration result for the user? | Yes, based on the provided information, it appears that the code is already designed to track requests, store module results, and display the orchestration result.
- The `request_id` is already included in the response.
- The `module results` from each module executed are also included in the response.
- The `orchest... |
{ "request_id": "53fc2dcd-399d-4a2b-8bde-912b9f001fed", "module_results": { "templating": [ { "role": "user", "content": "Reply with `Orchestration is Working!` in German" } ], "llm": { "id": "chatcmpl-9k8M3djXphXPWh2QkQm1YVtXK4Eki", "object": "chat.completion", "created": 1720782231, "model": "gpt-4o-mini", "choices":... | What is the German translation for "orchestration is working"? | The German translation for "Orchestration is Working" is "Orchestrierungsdienst funktioniert!". |
"prompt_tokens": 20, "total_tokens": 30 } } }, "orchestration_result": { "id": "chatcmpl-9k8M3djXphXPWh2QkQm1YVtXK4Eki", "object": "chat.completion", "created": 1720782231, "model": "<ModelName>", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "Orchestrierungsdienst funktioniert!" }, "finish_re... | Does the orchestration service work? | Yes, the orchestration service works, as indicated by the "choices" field in the "orchestration_result" object, which contains the message "Orchestrierungsdienst funktioniert!" (which translates to "Orchestration service works!"). |
10
INTERNAL
Generative AI Hub with SAP AI Core Generative AI Hub in SAP AI Core
}
The templating module result contains the user message with the filled in parameters. The LLM module result contains the response of the model execution. In this example, the LLM module result and the orchestration result are the same... | Can you provide an example where the LLM module result and the orchestration result in the Generative AI Hub with SAP AI Core Generative AI Hub differ due to the output filtering module? | Unfortunately, the provided snippet does not contain enough information to provide a clear example of where the LLM module result and the orchestration result differ due to the output filtering module.
However, based on the given information, here's a possible example:
Let's say the Generative AI Hub with SAP AI Core... |
Consumption of GenAI Models Using Orchestration – A Beginner's Guide
Libraries and SDKs
1.1.4 Next Steps
In the section above, we've outlined the basic steps for model consumption.
If you're interested in exploring advanced use cases for enhancing LLM responses with capabilities like grounding, data masking, and co... | What are some advanced use cases for enhancing LLM responses with capabilities like grounding, data masking, and content filtering? | Enhancing LLM responses with capabilities like grounding, data masking, and content filtering can be explored through Orchestration (see page 12). |
When to Use Orchestration vs. Foundation Models
Advantages of Using Orchestration
Provider Agnostic: The orchestration service offers a unified interface for accessing LLMs from multiple providers, simplifying the process of testing and comparing models. You can switch models via a config parameter without creating n... | What are the key benefits of using an orchestration service for Large Language Models (LLMs) integration in applications? | The key benefits of using an orchestration service for Large Language Models (LLMs) integration in applications are:
1. Provider Agnostic: A unified interface for accessing LLMs from multiple providers.
2. Simplified Integration: Streamlines the integration of LLMs into applications, ensuring efficient and cost-effect... |
Expandability: The service allows you to start with basic features and add modules as needed, offering a smooth learning curve. | Is the service expandable and easy to learn? | Yes, the service is expandable and has a smooth learning curve. |
Considerations Before Using Orchestration
Function Calling Availability: Function calling, with its structured output aspects, isn't currently available. • Open Source Framework Support: While we support LangChain, a popular open-source framework, we can't support every framework available. If a specific open-source f... | Can we extend the open-source framework for better Function calling availability in Orchestration? | Yes, according to the document, you might need to extend the open-source framework for better Function calling availability in Orchestration. |
INTERNAL
11
1.2 Accessing Generative AI Models Through Global
Scenarios
Access to generative AI models falls under the global AI scenarios foundation-models and orchestration. SAP AI Core manages these scenarios. You can access individual models as executables through serving templates. To use a specific model, cho... | What are the SAP AI Core scenarios for accessing individual models as executables through serving templates? | The SAP AI Core scenarios for accessing individual models as executables through serving templates are:
1. Global AI scenarios foundation-models
2. Global AI scenarios orchestration |
1.2.1 Orchestration
The orchestration service operates under the global AI scenario orchestration, which is managed by SAP AI Core. This service enables the use of various generative AI models with a unified code, configuration, and deployment.
In this orchestration, the harmonized API allows you to use different fou... | What are the key features of the orchestration service? | The key features of the orchestration service include:
1. Unified code, configuration, and deployment for various generative AI models.
2. The ability to use different foundation models without changing the client code, thanks to the harmonized API.
3. The ability to use different foundation models and versions, which... |
Templating: This feature lets you compose prompts with placeholders that are filled during inference. • Content filtering: This feature allows you to restrict the type of content passed to and received from a generative AI model.
Data masking: This feature enables data masking through anonymization or pseudonymization... | What are the key features of this AI model? | Templating, Content filtering, and Data masking are the key features of this AI model. |
Grounding: This feature lets you integrate external, contextually relevant, domain-specific, or real-time data into AI processes. This data enhances the natural language processing capabilities of pretrained models, which are trained on general material.
The following scenarios are available:
Global Scenario
Executa... | What are the available scenarios for integrating external data into AI processes? | The available scenarios for integrating external data into AI processes are:
1. Global Scenario
2. Executable ID |
aged by SAP AI Core.
12
INTERNAL
Generative AI Hub with SAP AI Core Generative AI Hub in SAP AI Core
1.2.1.1
Orchestration Workflow
In a basic orchestration scenario, you can combine different modules from orchestration into a pipeline that can be executed with a single API call. Within the pipeline, the response... | Can I design a basic orchestration workflow in SAP AI Core by combining different modules into a pipeline that can be executed with a single API call? | Yes, you can design a basic orchestration workflow in SAP AI Core by combining different modules into a pipeline that can be executed with a single API call. This is described in the document as a basic orchestration scenario. |
Prerequisites
You have created a deployment for orchestration as described at Create a Deployment for Orchestration in SAP AI Core [page 6]. | How do I create a deployment for orchestration in SAP AI Core as described in the documentation at page 6? | Based on the information provided, I will answer the question.
To create a deployment for orchestration in SAP AI Core as described in the documentation at page 6, follow these steps:
1. Create a deployment for orchestration in SAP AI Core.
2. This is described in the documentation at Create a Deployment for Orchestr... |
Workflow
The order in which the pipeline is executed is defined centrally in orchestration. However, you can configure the details for each module and omit optional modules by passing an orchestration configuration in JSON format with the request body.
The image below is interactive. Click each part of the workflow f... | What are the configuration options available for each module in the workflow pipeline? | The configuration options available for each module in the workflow pipeline are:
1. Grounding: No information provided in the given text.
2. Templating: No information provided in the given text.
3. Data Masking: No information provided in the given text.
4. Input Filtering: No information provided in the given text.... |
1.2.1.1.1
Grounding
Grounding integrates external, contextually relevant, domain-specific, or real-time data into AI processes. This data enhances the natural language processing capabilities of pretrained models, which are trained on general material. | What role does grounding play in enhancing the natural language processing capabilities of AI models? | Grounding integrates external, contextually relevant data into AI processes, which enhances the natural language processing capabilities of pretrained models by providing them with domain-specific information. |
Grounding is a service designed to handle data-related tasks, such as grounding and retrieval, using vector databases. It provides specialized data retrieval through these databases, grounding the retrieval process with your own external and context-relevant data. Grounding combines generative AI capabilities with the ... | What is the relationship between Generative AI Hub and SAP AI Core? | The Generative AI Hub is integrated with SAP AI Core. |
use real-time, precise data to improve decision-making and business operations for specific AI-driven business solutions.
Prerequisites
To use the Grounding module in the orchestration pipeline, you need to prepare the knowledge base in advance.
Generative AI hub offers multiple options for users to provide data (pr... | What kind of data repositories and pipeline options are supported by the Grounding module in the generative AI hub? | According to the document, the Grounding module in the generative AI hub supports the following:
1. Data repositories: Supported data repositories where users can upload documents.
2. Pipeline options: Running the data pipeline to vectorize the documents. |
Provide the chunks of document via Vector API directly. Provide the chunks of document via Vector API directly. For more information, see Vector API [page 25].
To use grounding, choose from one of the following options.
Option 1: Upload the documents to the supported data repository and run | What are the ways to provide document chunks via Vector API directly? | According to the document, there is only one way to provide document chunks via Vector API directly:
1. Upload the documents to the supported data repository and run. |
Data Pipeline
In this case, the pipeline collects documents and segments the data into chunks. It generates embeddings, which are multidimensional representations of textual information, and stores them efficiently in a vector database.
The process for providing unstructured documents with the Pipeline API looks like... | What are the initial steps required to set up the Data Pipeline for unstructured document processing? | The initial steps required to set up the Data Pipeline for unstructured document processing are:
1. Perform initial one-time administrative steps
2. Prepare vector knowledge base
3. Configure grounding module in the orchestration |
Hover over each action for a description. Click the action for more information.
Perform Initial One-Time Administrative Steps
Before you can prepare your data for the Pipeline API, you must create a resource group and a generic secret for grounding. For more information, see:
Create a Resource Group for Grounding [... | What are the initial one-time administrative steps required to prepare data for the Pipeline API? | To prepare data for the Pipeline API, the initial one-time administrative steps required are:
1. Create a resource group for grounding.
2. Create a generic secret for grounding.
These steps are necessary before preparing data for the Pipeline API, as described in the provided document. |
Prepare Vector Knowledge Base
You configure the Pipeline API to read unstructured data from data repositories and store it in a vector database. Use the Pipeline API to:
1. Read unstructured documents from various data repositories. Segment the data into chunks and generate
embeddings.
2. Store the multidimensional... | What are the key steps involved in configuring the Pipeline API to prepare a vector knowledge base from unstructured data? | Based on the provided information, the key steps involved in configuring the Pipeline API to prepare a vector knowledge base from unstructured data are:
1. **Read unstructured documents**: Use the Pipeline API to read unstructured data from various data repositories.
2. **Segment and generate embeddings**: Segment the... |
For more information, see Preparing Data Using the Pipeline API [page 23]. | What is Preparing Data Using the Pipeline API? | Preparing Data Using the Pipeline API is mentioned in the document, but the actual content about the Pipeline API is not provided in the given snippet. However, based on the general knowledge about the Pipeline API, it is a data processing framework that allows users to create a data pipeline, which is a series of data... |
Configure Grounding Module in the Orchestration
In the orchestration pipeline, you add configuration for the grounding requests:
1. Create a grounding request configuration in the orchestration pipeline using the repository IDs and filters.
2. Run the orchestration pipeline and check that the response refers to the ... | What is the correct configuration for the Grounding Module in the Orchestration pipeline? | The correct configuration for the Grounding Module in the Orchestration pipeline is to create a grounding request configuration using the repository IDs and filters. |
In this case, you provide chunks of data directly and store them using the Vector API.
The process for providing chunks of data with the Vector API looks like this:
Perform initial one-time administrative steps [page 15] • Prepare vector knowledge base [page 16] • Configure grounding module in the orchestration [page... | What is the process for providing chunks of data with the Vector API? | The process for providing chunks of data with the Vector API is as follows:
1. Perform initial administrative steps
2. Prepare a vector knowledge base
3. Configure grounding modules in the orchestration |
Perform Initial One-Time Administrative Steps
Before you can prepare your data for the Vector API, you must create a resource group. For more information, see Create a Resource Group for Grounding [page 16].
Note
When working with the Vector API, you do not need to create a generic secret for grounding.
Generativ... | What are the initial one-time administrative steps required to prepare data for the Vector API? | The initial one-time administrative steps required to prepare data for the Vector API are to create a resource group. For more information, see Create a Resource Group for Grounding [page 16]. |
Prepare Vector Knowledge Base
You provide chunks of information directly and store data in the vector database using the Vector API. Use the Vector API to:
1. Create collections.
2. Create documents by directly using the chunks of data provided by users.
3. Store data in the vector database.
4. Assign repository I... | Is it possible to use Vector API to prepare and store data in a vector database for future access? | Yes, it is possible to use the Vector API to prepare and store data in a vector database for future access. The Vector API can be used to create collections, create documents directly from user-provided chunks of information, store data in the vector database, and assign repository IDs to access the data. |
Configure Grounding Module in the Orchestration
In the orchestration pipeline, you add configuration for the grounding requests:
1. Create a grounding request configuration in the orchestration pipeline using the repository IDs and filters.
2. Run the orchestration pipeline and check that the response refers to the ... | What does the Grounding Module configuration look like in the Orchestration pipeline? | The process of configuring the Grounding Module in the Orchestration pipeline involves:
1. Creating a grounding request configuration using repository IDs and filters in the orchestration pipeline.
2. Running the orchestration pipeline and verifying that the response refers to the user data.
This process is described... |
Next Steps
Use the retrival API to search your data repositories.
For more information, see Retrieval API [page 34].
1.2.1.1.1.1 Administrative Tasks
1.2.1.1.1.1.1 Create a Resource Group for Grounding
Procedure
Create a resource group by sending a curl request, including the label that makes the resource group a... | What are the steps for creating a resource group that makes it available to the grounding service? | Based on the provided information, the steps for creating a resource group that makes it available to the grounding service are:
1. Use the Retrieval API to search data repositories (see page 34 for more information).
2. Create a resource group by sending a curl request, including the label that makes the resource gro... |
Note
Resource group Ids must be of length minimum: 3, maximum: 253. The first and last characters must be either a lowercase letter, an uppercase letter, or a number. Character entries from the second to penultimate, may include a lower case letter, an upper case letter, a number, a period (.), or a hyphen (-). No o... | What is the correct format for Resource group Ids in Azure? | The correct format for Resource group Ids in Azure is:
- Minimum length: 3 characters
- Maximum length: 253 characters
- The first and last characters must be either a lowercase letter, an uppercase letter, or a number
- The second to penultimate characters may include:
- Lowercase letters
- Uppercase letters
- ... |
curl --location --request POST "$AI_API_URL/v2/admin/resourceGroups" --header "Authorization: Bearer $TOKEN" \ --header 'Content-Type: application/json' \ --data-raw '{ "resourceGroupId": "{{resource_group}}", // Resource group ID to manage document grounding
16
INTERNAL
Generative AI Hub with SAP AI Core Generative... | What are the steps to manage a resource group with the ID "{{resource_group}}" using the Generative AI Hub in SAP AI Core? | The process to manage a resource group with the ID "{{resource_group}}" using the Generative AI Hub in SAP AI Core involves the following steps:
1. Send a POST request to the API endpoint "$AI_API_URL/v2/admin/resourceGroups".
2. Include the Authorization token in the request header with the key "Authorization" and th... |
"labels": [ // Labels for specifying resource purpose { "key": "ext.ai.sap.com/document-grounding", // Key for Document Grounding feature "value": "true" // Enable Document Grounding } ] }'
Existing resource groups can be made available to the grounding service using a PATCH request, including the document-grounding l... | What is the process for making an existing resource group available to the grounding service using a PATCH request with a document-grounding label? | The process for making an existing resource group available to the grounding service using a PATCH request with a document-grounding label is to include the "document-grounding" label in the PATCH request.
Here's a step-by-step breakdown:
1. Identify the existing resource group that needs to be made available to the... |
Procedure
1. Create an Application in Azure SAP SE tenant
a. Create a ServiceRequest for Microsoft Azure applications. For more information, see Service
Request
b. Log in to Entra and search for App registrations. For more information, see Entra
.
Under Owned Application you should be able to see the new applicat... | Can you give a summary of the process for creating and registering an application in Azure SAP SE tenant? | The process for creating and registering an application in Azure SAP SE tenant involves the following steps:
1. Create a Service Request for Microsoft Azure applications.
2. Log in to Entra and search for App registrations.
3. Once the application is created, select it and navigate to the API Permissions tab.
This pr... |
d. Navigate to Add permission Microsoft Graph Application Permissions .
e. Search Sites.Selected and select it.
f. Navigate to the Certificates and Secrets tab and then to Client Secret New Client Secret
.
g. Give a description and expiry date of your choice, and add your secret.
h. Navigate to the Overview tab an... | Are you setting up a SharePoint site with SAP SE and granting necessary permissions through Microsoft Graph Application Permissions? | Yes, the process described in the document is setting up a SharePoint site with SAP SE and granting necessary permissions through Microsoft Graph Application Permissions. |
a. Use the Navigator to create a new share point site. For more information, see Navigator
b. Add your self as a “Responsible User” for the site.
c. Create a ServiceRequest for the site administrator role. For more information, see ServiceRequesr
and Sample Request
.
Generative AI Hub with SAP AI Core Generative A... | Will the instructions in a. and b. be applied to the creation of a new share point site in the context of the Generative AI Hub with SAP AI Core? | Based on the provided information, it appears that the instructions in a. and b. are general steps for creating a new SharePoint site, but there is no explicit mention of the Generative AI Hub with SAP AI Core. However, since the instructions are mentioned in the context of the Generative AI Hub with SAP AI Core, it ca... |
You will need to include the following details on your ticket: • Application Name :- <Name of the above Azure application> • Site URL :- <Full SharePoint site URL> • Site Name :- <Your new SharePoint Site name> • Application (Client) ID :- <ApplicationId of the above Azure application> • Display Name :- <Any name of yo... | What details should be included on a ticket for the Azure application? | The process to include on the ticket is as follows:
• Application Name: <Name of the Azure application>
• Site URL: <Full SharePoint site URL>
• Site Name: <New SharePoint Site name>
• Application (Client) ID: <ApplicationId of the Azure application>
• Display Name: <Any chosen name>
• Add this line: Please provide th... |
permission for the given SPN in the SE tenant . Also provide "Site Collection Administrator" permission for my user as I am the admin for this site.
d.
In Microsoft SharePoint, navigate to Site Access Advanced Settings . Site Collection Administrator should be enabled.
e.
In Microsoft Entra, navigate to <Your Appli... | Can you please provide the permission for the given SPN in the SE tenant, as well as grant "Site Collection Administrator" permission for my user since I am the site admin? | Based on the provided information, the process to grant the permission for the given SPN in the SE tenant and grant "Site Collection Administrator" permission for your user as the site admin is as follows:
1. In Microsoft SharePoint, navigate to Site Access Advanced Settings.
2. Enable Site Collection Administrator.
3... |
a.
Install Microsoft PowerShell app on your local computer. For more information, see Installing Microsoft PowerShell
.
b.
Install the PnP PowerShell Module. For more information, see Installing PnP PowerShell PowerShell
.
and PnP | Can you install Microsoft PowerShell on your local computer for more information see Installing Microsoft PowerShell? | Yes, you can install Microsoft PowerShell on your local computer. For more information, see the link provided in the document: "Installing Microsoft PowerShell". |
Note
The latest version of the PnP PowerShell module (2.x) requires PowerShell version 7.2 or later. For more information on installing PowerShell 7.x on Windows, Linux, and Mac devices, see Installing Microsoft PowerShell
.
c. Request access to the PnP Management Shell. For more information, see My Access
.
d. ... | Is the PnP PowerShell module compatible with PowerShell version 7.2 or later? | Yes, the PnP PowerShell module is compatible with PowerShell version 7.2 or later, as stated in the document: "The latest version of the PnP PowerShell module (2.x) requires PowerShell version 7.2 or later." |
93ccfefd-a4cb-4e85-84d1-9cbacd93022a
2. Grant-PnPAzureADAppSitePermission -AppId "<your app Id>" -DisplayName
"<Any name of your choice>" -Site <complete site Url> -Permissions Write
You should receive a response containing Roles: {Write}
Related Information
Use Sites.Selected for accessing SharePoint Site | What are the required parameters to configure Grant-PnPAzureADAppSitePermission for Azure AD application permissions? | Based on the provided information, the required parameters to configure Grant-PnPAzureADAppSitePermission for Azure AD application permissions are:
1. -AppId: The ID of your Azure AD application.
2. -DisplayName: A display name of your choice for the permission.
3. -Site: The complete URL of the SharePoint site.
4. -P... |
1.2.1.1.1.1.3 Generic Secrets for Grounding
The following data repositories are supported in the grounding module:
18
INTERNAL
Generative AI Hub with SAP AI Core Generative AI Hub in SAP AI Core
Microsoft SharePoint
The following authentication types for each document repository type:
Microsoft SharePoint • OAut... | What authentication types are supported for Microsoft SharePoint repositories in the grounding module? | OAuth2Password and OAuth2ClientCredentials are the supported authentication types for Microsoft SharePoint repositories in the grounding module. |
1.2.1.1.1.1.3.1 Grounding Generic Secrets for Microsoft
SharePoint
Prerequisites
You have prepared your SharePoint integration. For more information, see Prepare your Microsoft SharePoint Integration [page 17].
You have created a resource group for grounding. For more information, see Create a Resource Group for Gr... | What are the prerequisites for implementing Microsoft SharePoint grounding? | The prerequisites for implementing Microsoft SharePoint grounding are:
1. You have prepared your SharePoint integration.
2. You have created a resource group for grounding. |
Procedure
Send a POST request and enter the URL {{apiurl}}/v2/admin/secrets. • AI-Tenant-Scope : true. The operation will be performed at the main tenant level. • AI-Resource-Group : <resource-group-name>. The operation will be performed at the resource- | What is the API endpoint URL to send a POST request for adding a secret in the main tenant level with a specific resource group? | The API endpoint URL to send a POST request for adding a secret in the main tenant level with a specific resource group is:
{{apiurl}}/v2/admin/secrets
This URL is specified in the procedure, and the additional parameters AI-Tenant-Scope and AI-Resource-Group are used to specify the main tenant level and the resource... |
group level.
In your request, you'll need to specify the resource group you created for grounding purposes, and set the document grounding label to true.
The following API request is used to create a generic secret with base64-encoded values, such as client credentials and other necessary fields, for integrating grou... | Is it necessary to specify the resource group for grounding purposes when creating a generic secret for the generative AI hub? | Based on the provided information, it appears that specifying the resource group for grounding purposes is necessary when creating a generic secret for the generative AI hub. This is because the API request mentions that you'll need to specify the resource group you created for grounding purposes. |
Note
In API permissions: • If you configure access from SAP BTP with OAuth2Password authentication, use the permission name
Sites.Read.All with type Delegated
If you configure access from SAP BTP with OAuth2ClientCredentials authentication, use the permission name Sites.Selected with type Application
Generative A... | Does the permission name Sites.Read.All have a type of Delegated for OAuth2Password authentication in API permissions? | Yes, the permission name Sites.Read.All has a type of Delegated for OAuth2Password authentication in API permissions. |
Example with OAuth2Password:
Populate the example codeblock with the following information:
The secret name is written without hyphens to make it simple to consume as a Unix environment variable later. It is written in capitals as is convention.
Field
Value
Name
Name of the generic secret to be created
Descripti... | What is the conventional method of representing a Unix environment variable name in an OAuth2Password example? | The conventional method of representing a Unix environment variable name in an OAuth2Password example is to write it in capitals without hyphens. |
Authentication
Base64 encoded value for OAuth2Password
User
Base64 encoded value for your Microsoft SharePoint techni-
cal user
Password
Base64 encoded value for your Microsoft SharePoint pass-
word
Client ID
Base64 encoded value for your Microsoft SharePoint Appli-
cation credentials
Client Secret
Base64 e... | What is the Base64 encoded value for OAuth2Password? | The Base64 encoded value for OAuth2Password is Authentication. |
Token Service URL
Base64 encoded value for https://login.microsoftonline.com/
<tenant-id>/oauth2/v2.0/token
Type
Base64 encoded value for HTTP
Proxy Type
Base64 encoded value for Internet
Token Service URL Type
Base64 encoded value for Dedicated
curl --location --request POST "$AI_API_URL/v2/admin/secrets" \ -... | What is the Base64 encoded value for https://login.microsoftonline.com/ <tenant-id>/oauth2/v2.0/token? | The Base64 encoded value for https://login.microsoftonline.com/<tenant-id>/oauth2/v2.0/token is the Token Service URL. |
--header 'Content-Type: application/json' \ --header 'AI-Resource-Group: <resource group created for grounding>' \ --data-raw '{ "name": "<generic secret name>", "data": { "description": "<description of generic secret>", "clientId": "<client id>", "authentication": "T0F1dGgyUGFzc3dvcmQ=", "tokenServiceURL": "<token se... | What is the structure and format of the JSON data being sent with the POST request to create a generic secret? | The structure and format of the JSON data being sent with the POST request to create a generic secret is as follows:
- The JSON data is enclosed within curly brackets `{}`.
- It contains the following key-value pairs:
- `"name": "<generic secret name>"` - a string representing the name of the generic secret.
- `"d... |
"scope": "aHR0cHM6Ly9ncmFwaC5taWNyb3NvZnQuY29tLy5kZWZhdWx0", "type": "SFRUUA==", "proxyType": "SW50ZXJuZXQ=" }, "labels": [ { "key": "ext.ai.sap.com/document-grounding", // Label for Document Grounding feature "value": "true" }, { "key": "ext.ai.sap.com/documentRepositoryType", // Label for Document Repository Type "va... | What is the value of the "proxyType" in the JSON object? | The value of the "proxyType" in the JSON object is "SW50ZXJuZXQ=". However, this seems to be a base64 encoded string. When decoded, it becomes "https://". |
Example with OAuth2ClientCredentials:
Populate the example codeblock with the following information:
The secret name is written without hyphens to make it simple to consume as a Unix environment variable later. It is written in capitals as is convention.
Field
Value
Name
Name of the generic secret to be created
... | What information do I need to populate the example codeblock for OAuth2ClientCredentials? | To populate the example codeblock for OAuth2ClientCredentials, you need the following information:
1. Name of the generic secret to be created (in capitals, without hyphens)
2. Base64 encoded value for the description of the generic secret to be created
3. Base64 encoded value for the URL https://sap.sharepoint.com/si... |
Authentication
Base64 encoded value for OAuth2ClientCredentials
Client ID
Base64 encoded value for your Microsoft SharePoint Appli-
cation credentials
Client Secret
Base64 encoded value for your Microsoft SharePoint Appli-
cation credentials
Generative AI Hub with SAP AI Core Generative AI Hub in SAP AI Core
I... | What is the base64 encoded value for OAuth2ClientCredentials in authentication? | The base64 encoded value for OAuth2ClientCredentials in authentication is not explicitly mentioned in the provided information. However, it is mentioned that "Base64 encoded value for OAuth2ClientCredentials" is listed as "Authentication" in the table. |
Token Service URL
Base64 encoded value for https://login.microsoftonline.com/ <tenant-id>/oauth2/v2.0/token
Type
Base64 encoded value for HTTP
Proxy Type
Base64 encoded value for Internet
Token Service URL Type
Base64 encoded value for Dedicated
User Name
Empty String
Password
Empty String | What is the Base64 encoded value for https://login.microsoftonline.com/<tenant-id>/oauth2/v2.0/token? | The process is described as "Token Service URL".
The Base64 encoded value for https://login.microsoftonline.com/<tenant-id>/oauth2/v2.0/token is the Token Service URL. |
Sample Code
curl --request POST \ --url $AI_API_URL/v2/admin/secrets \ --header 'AI-Resource-Group: <resource group created for grounding>' \ --data-raw '{ "name": "<generic secret name>", "data": { "description": "<description of generic secret>", "clientId": "<client id>", "authentication": "T0F1dGgyQ2xpZW50Q3JlZG... | What type of API call is being made in the provided sample code? | The type of API call being made in the provided sample code is a POST request. |
"clientSecret": "<client secret>", "scope": "aHR0cHM6Ly9ncmFwaC5taWNyb3NvZnQuY29tLy5kZWZhdWx0", "type": "SFRUUA==", "proxyType": "SW50ZXJuZXQ=", "user": "", "password": "" }, }, "labels": [ { "key": "ext.ai.sap.com/document-grounding", // Label for Document Grounding feature "value": "true" }, { "key": "ext.ai.sap.com/... | What is the secret key used for authentication in this Generative AI Hub configuration in SAP AI Core? | The secret key used for authentication in this Generative AI Hub configuration in SAP AI Core is the "clientSecret" which is "<client secret>". |
Related Information
Update a Generic Secret Delete a Generic Secret | Can you update a generic secret? Can you delete a generic secret? | Based on the provided information, it appears to be a list of related actions:
1. Update a Generic Secret
2. Delete a Generic Secret
From this, we can infer that:
- Yes, you can update a generic secret.
- Yes, you can delete a generic secret. |
1.2.1.1.1.2 Data Preparation
1.2.1.1.1.2.1 Pipeline API
The Pipeline API allows you to provide unstructured documents from your data repositories.
The API is proxied through the generative AI hub and incorporates vector stores, such as the managed HANA database. This pipeline segments data into chunks and generates ... | How does the Pipeline API segment data in the generative AI hub? | According to the document, the Pipeline API segments data into chunks and generates embeddings, which are multidimensional representations of textual information, through the generative AI hub. |
The Pipeline API is compatible with the following data repositories:
Microsoft SharePoint:
Maximum of 2000 documents • PDF, DOCX file types • Plain text format • Content is updated once each day
SAP Help Portal: Content available at http://help.sap.com is supported.
For region and pricing information, see SAP Note ... | What types of data repositories is the Pipeline API compatible with? | The Pipeline API is compatible with the following data repositories:
1. Microsoft SharePoint
2. SAP Help Portal (content available at http://help.sap.com) |
Prerequisites
You have created a resource group (see Create a Resource Group for Grounding [page 16]). • You have created a generic secret (see Grounding Generic Secrets for Microsoft SharePoint [page 19]).
Context
Generative AI Hub with SAP AI Core Generative AI Hub in SAP AI Core
INTERNAL
23
Procedure
The foll... | What are the prerequisites for using the Generative AI Hub with SAP AI Core? | The prerequisites for using the Generative AI Hub with SAP AI Core are:
1. You have created a resource group.
2. You have created a generic secret. |
Sample Code
curl --request POST \ --url $AI_API_URL/v2/lm/document-grounding/pipelines \ --header 'AI-Resource-Group: {{resource_group}}' \ --header 'Content-Type: application/json' \ --data '{ "type": "MSSharePoint", "configuration": { "destination": "<generic secret name>", "sharePoint": { "site": { "name": "<site... | Does the provided code snippet utilize the `curl` command to create a POST request to the Document Grounding API for MS SharePoint? | Yes, the provided code snippet utilizes the `curl` command to create a POST request to the Document Grounding API for MS SharePoint. |
curl --request POST \ --url $AI_API_URL/v2/lm/document-grounding/pipelines \ --header 'AI-Resource-Group: {{resource_group}}' \ --header 'Content-Type: application/json' \ --data '{ "type": "MSSharePoint", // Currently, only Microsoft SharePoint is supported "configuration": { "destination": "<generic secret name>", //... | Is the `curl` command used to send a POST request to the AI API endpoint for document grounding pipelines? | Yes, the `curl` command is used to send a POST request to the AI API endpoint for document grounding pipelines. |
"<folder2>"] // [Optional] Array of folders within the SharePoint site for selective indexing } } } }' | What are the folders within the SharePoint site that can be selectively indexed? | The folders within the SharePoint site that can be selectively indexed are specified in the "[<folder2>]" array. |
YAML Metadata Warning:The task_categories "text2text-generation" is not in the official list: text-classification, token-classification, table-question-answering, question-answering, zero-shot-classification, translation, summarization, feature-extraction, text-generation, fill-mask, sentence-similarity, text-to-speech, text-to-audio, automatic-speech-recognition, audio-to-audio, audio-classification, audio-text-to-text, voice-activity-detection, depth-estimation, image-classification, object-detection, image-segmentation, text-to-image, image-to-text, image-to-image, image-to-video, unconditional-image-generation, video-classification, reinforcement-learning, robotics, tabular-classification, tabular-regression, tabular-to-text, table-to-text, multiple-choice, text-ranking, text-retrieval, time-series-forecasting, text-to-video, image-text-to-text, image-text-to-image, image-text-to-video, visual-question-answering, document-question-answering, zero-shot-image-classification, graph-ml, mask-generation, zero-shot-object-detection, text-to-3d, image-to-3d, image-feature-extraction, video-text-to-text, keypoint-detection, visual-document-retrieval, any-to-any, video-to-video, other
Dataset Card for genaihub-dataset
This dataset has been created with distilabel.
Dataset Summary
This dataset contains a pipeline.yaml which can be used to reproduce the pipeline that generated it in distilabel using the distilabel CLI:
distilabel pipeline run --config "https://huggingface.co/datasets/allurisravanth/genaihub-dataset/raw/main/pipeline.yaml"
or explore the configuration:
distilabel pipeline info --config "https://huggingface.co/datasets/allurisravanth/genaihub-dataset/raw/main/pipeline.yaml"
Dataset structure
The examples have the following structure per configuration:
Configuration: default
{
"context": "d e v r e s e r s t h g i r\n\nl l\n\nA\n\n. y n a p m o c e t a\n\ni l\n\nffi a P A S n a r o E S P A S 5 2 0 2 \u00a9\n\nINTERNAL\n\n2025-02-25\n\nGenerative AI Hub with SAP AI Core\n\nTHE BEST RUN",
"question": "What is the date of the recent update (2025-02-25) related to the Generative AI Hub with SAP AI Core?",
"response": "The date of the recent update is 2025-02-25."
}
This subset can be loaded as:
from datasets import load_dataset
ds = load_dataset("allurisravanth/genaihub-dataset", "default")
Or simply as it follows, since there's only one configuration and is named default:
from datasets import load_dataset
ds = load_dataset("allurisravanth/genaihub-dataset")
- Downloads last month
- 5