--- title: "Prompt AI" slug: "chat-completion" updated: 2026-10-01T17:40:32Z published: 2026-10-01T17:40:32Z canonical: "documentation.decisions.com/chat-completion" --- > ## Documentation Index > Fetch the complete documentation index at: https://documentation.decisions.com/llms.txt > Use this file to discover all available pages before exploring further. # Prompt AI | Step Details | | --- | | **Introduced in Version** | 9.3.0 | | **Last Modified in Version** | 10.3 | | **Step Location(s)** | **AI > Anthropic > Prompt AI** **AI > AWS Bedrock** **AI > Google Gemini > Prompt AI** **AI > Azure Foundry** **AI > Ollama** **AI > OpenAI> Prompt AI** **AI > Common > Prompt AI** | The Prompt AI Step is used to generate a Chat Response and is available for use across several AI Modules. The step enables Users to input Model and AI Prompt information to output a Chat Response. To utilize this step, Users must have AI Common and the desired AI Module installed, as these Modules will be used to populate the Chat Response. Since this step is available across several AI Modules, Users can quickly switch between models without an additional step. ### Prerequisites - This step requires the [AI Common Module](https://documentation.decisions.com/docs/ai-common) to be [installed](https://documentation.decisions.com/docs/installing-modules-decisions) before it will be available in the toolbox. - Users must create a [Project Dependency](https://documentation.decisions.com/docs/project-dependencies) for the AI Common Module and any other AI Modules that will use the Prompt AI step. - Users need to be on v9 to access this step. --- ## Properties #### Settings | Property | Description | | --- | --- | | Override Default Provider | This setting is available when utilizing the step through AI Common. Checking this box enables Users to override the default provider (AI Common) and choose another AI Module, such as Google Gemini. | | Override API Key **(v9.13+)** | Instead of using the key set at the System level on the module configuration, enabling this setting allows Users to add another key. | | Get Provider From Flow | Enabling this setting allows the AI provider to be picked from Flow data, rather than hard-coded at the Step level. This allows for dynamic provider shifting. | | Select Model | This is a checkbox that allows you to select a custom Model for the step to use. | | Model | This field is a drop-down menu that appears when Users click the **Select Model** checkbox. The drop-down menu will display a choice between available models depending on which module the step is pulled from. The step will search and display models depending on which AI Modules are installed in the Decisions environment. | | Get Model From Flow | This setting will allow you to dynamically select a model for the step to use. By checking this, you will allow the step to have a new string input called **Model.** | | Select Prompt | By checking this checkbox, you are choosing to select a predesigned Prompt as opposed to writing and managing your prompt here in the flow. | | Get Prompt from Flow | This field will appear if a User clicks the **Select Prompt** checkbox. This enables Users to select a prompt using the Flow. | | Use TextMerge Prompt | This field will appear if a User clicks the **Select Prompt** checkbox. Clicking the plus icon on the right-hand side of the drop-down menu populates a form to select a prompt built using the [AI Prompt Manager](https://documentation.decisions.com/v9/docs/prompt-manager-1). | | Prompt | If Users do not check any of the above check boxes, a prompt can be manually entered in the Prompt field. | #### Advanced | Property | Description | Data Type | | --- | --- | --- | | Attachments | Allows Users to attach images or PDF files as multimodal input to the AI model. | -- | | Generation Parameters **(For v9.32+, and v10.3+)** | Allows Users to specify/edit the following parameters. - Temperature - Top P (nucleus sampling) - Max Tokens - Stop Sequence - Top K - Candidate Count - Seed Each LLM has different supported Generation Parameters. | -- | | Message History | Allows Users to specify a history of prior messages. | -- | | Response Schema | Allows Users to specify a data type to enforce the structured JSON output from the AI model. | -- | | Tool Flow | Allows Users to specify an AI Tool Flow. | -- | | Web Search **(For v9.32+, and v10.3+)** | Allows Users to enable/disable web search. | Boolean | #### Inputs | Property | Description | Data Type | | --- | --- | --- | | Override Variables | Users can build an Array on this step to expose the **Find** and **Replace** fields. Users can then designate variables in both fields to see customized results in the Chat Response Output. | ListofAIPromptVariables | #### Outputs | Property | Description | Data Type | | --- | --- | --- | | Chat Response | A list of AI Prompt AI utilizing the selected model, prompt, and override variables if applicable. | List of AIPromptAI | --- ## Common Errors ### Using a Bad API Key When utilizing the Prompt AI Step with the [OpenAI](https://documentation.decisions.com/docs/openai) Module, the step will not return desired results unless the [API Key](https://documentation.decisions.com/docs/openai?highlight=Open%20AI#generate-openai-api-keys) is working properly. If an incorrect API Key is entered on the **OpenAI Settings Page**, the Step will error out in the Debugger. To correct this, ensure the API Key utilized is accurate, and there are no issues with an associated OpenAI account. ### Exception Message: ```csharp Exception Stack Trace: DecisionsFramework.Design.Flow.ErrorRunningFlowStep: Error running step Prompt AI 2[CreatePromptAIStep] in flow [Using an AI Prompt in a Flow]: HTTP 429 (insufficient_quota: insufficient_quota) You exceeded your current quota, please check your plan and billing details. For more information on this error, read the docs: https://platform.openai.com/docs/guides/error-codes/api-errors. ---> System.ClientModel.ClientResultException: HTTP 429 (insufficient_quota: insufficient_quota) You exceeded your current quota, please check your plan and billing details. For more information on this error, read the docs: https://platform.openai.com/docs/guides/error-codes/api-errors. at OpenAI.ClientPipelineExtensions.ProcessMessage(ClientPipeline pipeline, PipelineMessage message, RequestOptions options) at OpenAI.Chat.ChatClient.CompleteChat(BinaryContent content, RequestOptions options) at OpenAI.Chat.ChatClient.CompleteChat(IEnumerable`1 messages, ChatCompletionOptions options, CancellationToken cancellationToken) at OpenAI.Chat.ChatClient.CompleteChat(ChatMessage[] messages) at Decisions.AI.OpenAI.Steps.Completions.CreateChatCompletionStep.CreateChatCompletion(CreateChatCompletionRequest request) at Decisions.AI.OpenAI.Steps.Completions.CreateChatCompletionStep.Run(StepStartData data) at DecisionsFramework.Design.Flow.FlowStep.RunStepInternal(String flowTrackingID, String stepTrackingID, KeyValuePairDataStructure[] stepRunDataValues, AbstractFlowTrackingData trackingData) at DecisionsFramework.Design.Flow.FlowStep.Start(String flowTrackingID, String stepTrackingID, FlowStateData data, AbstractFlowTrackingData trackingData, RunningStepData currentStepData) --- End of inner exception stack trace --- ``` --- ### Using the Prompt AI Step in a Flow The following example demonstrates utilizing the Prompt AI step with a Model from Common. To set up the Step in Decisions: 1. Drag the **Prompt AI** step onto the workspace and connect it to the **Start** and **End** steps. 2. Navigate to the Properties Panel and customize the settings according to the desired Model. In this example, Gemini -2.0-flash was used as the model, and a prompt was selected from the AI Manager. 3. For the **Inputs**, enter desired variables into the **Find** and **Replace** fields. Ensure these variables correspond to the chosen prompt. ![](https://cdn.document360.io/6ef8bcc1-6489-4486-9ad1-83acff7e5df0/Images/Documentation/AICommon%20Prompt%20AI%20Default%20Model(1).png) After the desired settings and **Inputs** have been configured in the Properties Panel, select **Debug.** The Flow should generate a **Chat Response** that represents something similar to the screenshot below. #### ![](https://cdn.document360.io/6ef8bcc1-6489-4486-9ad1-83acff7e5df0/Images/Documentation/2024-12-30_15h47_23(1).png) --- ## Step Changes | Description | Version | Release Date | Developer Task | | --- | --- | --- | --- | | Added Web Search ability to Prompt AI and Agent steps. Supported providers: OpenAI, Gemini, Anthropic, Foundry, LiteLLM(Anthropic, Gemini, OpenAI tested). Bedrock and Ollama are not supported. | 9.32, 10.3 | October 2026 | [DT-047702] | | Added Generation Parameters for Prompt AI steps. - Temperature - Top P (nucleus sampling) - Max Tokens - Stop Sequence - Top K - Candidate Count - Seed Each LLM has different supported Generation Parameters. | 9.32, 10.3 | October 2026 | [DT-044659] | | Chat completion steps now output the token count (input/output/total) for each provider with AI Common. | 10.2 | September 2026 | [DT-046939] | | **Prompt AI** steps now automatically use the configured AI default model for the selected provider, including Project Defaults and System Defaults. Users can still override the default model when needed, revert back to the configured default, and rely on the correct model being used during Flow execution, including after provider default model changes, removed defaults, and upgrades. | 10.2 | September 2026 | [DT-047744] | | The **Chat Completion** Step has been renamed to **Prompt AI**. Existing chat completion steps will be renamed to Prompt AI. | 10.2 | September 2026 | [DT-047769] | | Added the Override API Key setting. | [9.13](https://documentation.decisions.com/v99/docs/version-913x-release-notes) | July 2025 | [DT-044658] | --- Related Information [Utilizing AI Modules in Decisions](https://documentation.decisions.com/docs/utilizing-ai-modules-in-decisions)