> For the complete documentation index, see [llms.txt](https://docs.ovaledge.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.ovaledge.com/release8.3/ai/ai-prompt/ai-model-configuration.md). # AI Model Configuration AI Prompt and other supported AI capabilities use the AI provider and models configured in **Administration > System Settings > AI**. To configure the AI provider and models: 1. Go to **Administration**. 2. Select **System Settings**. 3. Open the **AI** tab. 4. Search for the **ai.config** system setting. 5. Select the configured value to open the **AI Model Configuration** window. The AI Model Configuration window supports the following providers: * OpenAI Cloud API * Gemini Cloud API * Anthropic Cloud API * Amazon Bedrock Cloud API {% hint style="info" %} The available models and configuration fields depend on the selected AI provider. {% endhint %} ## Model Configuration The following model configuration fields are available for the selected AI provider: ![](https://content.gitbook.com/content/8fgltewNgYZ8ucAp7Svt/blobs/m1b4GtJxzmemTZpX4FSF/unknown.png) | Field | Description | | -------------------- | ---------------------------------------------------------------------------------- | | API Key | Specifies the API key required to authenticate with the selected AI provider. | | Classification Model | Specifies the model used to classify and process supported AI requests. | | Summarization Model | Specifies the model used to generate summaries for supported AI requests. | | Large Context Model | Specifies the model used when processing requests that require a larger context. | | Reasoning Model | Specifies the model used for requests that require deeper or multi-step reasoning. | The available models in each list depend on the selected AI provider. **Example**: When OpenAI is selected, OpenAI models can be assigned to the Classification, Summarization, Large Context, and Reasoning Model fields. When Gemini, Anthropic, or Amazon Bedrock is selected, the corresponding supported models are available for selection. ### OpenAI Configuration When OpenAI Cloud API is selected, provide the API key and select the required models for the supported AI processing requirements. The Advanced Settings include: | Field | Description | | --------------- | ------------------------------------------------------------------------------ | | Base URL | Specifies the OpenAI endpoint or gateway URL. | | Timeout | Defines the maximum time allowed for an AI request. | | Max Tokens | Defines the maximum number of tokens allowed for the request. | | Retry Count | Defines how many times the system retries a failed request. | | Embedding Model | Specifies the model used to generate embeddings for supported AI capabilities. | ### Gemini Configuration When Gemini Cloud API is selected, provide the API key and select the required Gemini models. The Advanced Settings section includes the following fields: | Field | Description | | --------------- | ------------------------------------------------------------------------------------- | | Timeout | Specifies the maximum time allowed for an AI request before the request times out. | | Max Tokens | Specifies the maximum number of tokens allowed in the model response. | | Retry Count | Specifies the number of times the application retries a failed AI request. | | Embedding Model | Specifies the Gemini model used to generate embeddings for supported AI capabilities. | **Example**: The Embedding Model list can contain a Gemini embedding model such as gemini-embedding-001, depending on the available models for the configured provider. ### Anthropic Configuration When Anthropic Cloud API is selected, provide the API key and select the required Anthropic models. The Advanced Settings section includes the following fields: | Field | Description | | -------------- | ------------------------------------------------------------------------------------------------------------------------------------------------ | | Base URL | Specifies the base URL used to connect to the Anthropic service. | | Timeout | Specifies the maximum time allowed for an AI request before the request times out. | | Temperature | Controls the variation in the generated AI response. | | Top P | Controls the range of probable responses considered by the model during generation. | | Top K | Specifies the number of highest-probability tokens considered during response generation. | | Max Tokens | Specifies the maximum number of tokens allowed in the model response. | | Retry Count | Specifies the number of times the application retries a failed AI request. | | Stop Sequences | Specifies one or more sequences that instruct the model to stop generating a response. Multiple values can be entered as comma-separated values. | ### Amazon Bedrock Configuration When Amazon Bedrock Cloud API is selected, select the required models and configure the AWS connection and runtime settings. The Advanced Settings section includes the following fields: | Field | Description | | --------------------- | ----------------------------------------------------------- | | AWS Region | Selects the AWS Region used for the AI service. | | AWS Access Key ID | Specifies the AWS access key. | | AWS Secret Access Key | Specifies the AWS secret access key. | | AWS Session Token | Provides an optional temporary AWS session token. | | AWS Profile | Specifies the AWS profile to use. | | Assume Role ARN | Specifies the IAM role that OvalEdge can assume. | | Role Session Name | Defines the session name used when assuming the role. | | Timeout | Defines the maximum time allowed for an AI request. | | Temperature | Controls the variability of generated responses. | | Top P | Controls token selection based on cumulative probability. | | Max Tokens | Defines the maximum number of tokens allowed for a request. | | Retry Count | Defines the number of retry attempts. | | Rerank Region | Selects the AWS Region used for the rerank model. | ### Embedding Model The Embedding Model field specifies the model used to convert text into numerical representations, called embeddings. OvalEdge uses embeddings for supported embedding-based AI capabilities, such as semantic search and other AI-driven processing. The available embedding models depend on the selected AI provider. The Embedding Model field is displayed within the provider configuration. Therefore, its available options vary according to the selected provider. **Example**: * OpenAI can use an OpenAI embedding model. * Gemini can use a Gemini embedding model. * Amazon Bedrock can use a supported Bedrock embedding model, such as a supported Amazon Titan embedding model. ### Save the Configuration After selecting the provider, models, and required provider-specific settings: 1. Review the configured values. 2. Select **Save**. 3. Reopen the **AI Model Configuration window**, if required. 4. Verify that the selected provider and model configurations are retained. *** Copyright © 2025, OvalEdge LLC, Peachtree Corners, GA, USA. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.ovaledge.com/release8.3/ai/ai-prompt/ai-model-configuration.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.