Configuring the Chat Agent Innovation Release

The EDB Hybrid Manager (HM) Chat Agent is delivered with HM and powered by a chat-completion model that you provide through an HM-managed inference service. As an HM administrator, you create or register an inference service for that model. After that, anyone in your organization can use Chat Agent from the HM console.

This page is for HM administrators. To learn how to use Chat Agent, see the overview.

Before you begin

Confirm the following:

• HM is installed and reachable.
• The AI scenario is enabled in your HM deployment. Chat Agent is part of the AI scenario, and the sparkle icon stays hidden when the AI scenario isn't installed. To enable it, make sure ai is listed under scenarios in your HybridControlPlane manifest. See HM installation scenarios for details.
• HM AI Factory is available, with Model Serving and the Model Library reachable from the HM console. See Model Serving and Model Library for the AI Factory documentation.
• You have the AI Model Manager organization role. This role allows you to register AI models, apply model functions, and create inference services. See Roles and permissions for the full role matrix.
• You have the credentials and connection details for at least one chat-completion model — either a third-party API or a self-hosted endpoint. See Supported model providers.
• The chat-completion model supports a context window of at least 128K tokens. Chat Agent loads skill instructions, tool schemas, and conversation history on every turn — models with smaller context windows don't have enough context.

Configuring the model

Chat Agent is powered by a chat-completion model served through an HM-managed inference service. To make a model available to Chat Agent, go to Estate > Inference Services in the HM console, open the Quick Actions menu, and choose one of the following two paths depending on where your model runs:

• Create Local Inference Service — deploy and serve a model from your HM local Asset Library. Recommended when you want to keep prompts and responses inside your infrastructure.
• Register External Inference Service — point HM to a model endpoint running outside your cluster. Examples include a third-party cloud API (OpenAI, Anthropic, Gemini) or a self-hosted endpoint on another HM cluster. Recommended when you already operate the model elsewhere and want HM to use it.

Creating a local inference service

Add a model to your HM Asset Library, then create the inference service from that entry.

1. In Asset Library > Models, select + Add New Model and fill in the model details. Add chatbot-gen-content to the Functions field. For the full form reference, see How to deploy AI Models from the Model Library.
2. Go to Estate > Inference Services.
3. From Quick Actions, select Create Local Inference Service (or the Create Local Inference Service button in the empty state).
4. Pick the model you added in the previous step and complete the deployment form.
5. Submit and wait until the inference service reaches the Ready state.
6. (Optional) To customize Chat Agent behavior for this model, apply additional model functions. See Configuring model functions.

The next time a user opens Chat Agent, the model is listed in the model selector.

Registering an external inference service

Point HM to a model endpoint running outside your cluster — such as OpenAI's API, an Anthropic-compatible endpoint, a Google Gemini endpoint, or a self-hosted endpoint on another HM cluster.

1. Go to Estate > Inference Services.
2. From Quick Actions, select Register External Inference Service.
3. Provide the connection details for the external endpoint. Required fields are the URL, the API protocol (OpenAI v1, Anthropic Messages, or Google Gemini), and API key.
4. Confirm the chatbot-gen-content function is applied.
5. Submit and wait until the registration reaches the Ready state.
6. (Optional) To customize Chat Agent behavior for this model, apply additional model functions. See Configuring model functions.

For more on external services, see Model Serving in the AI Factory documentation.

Configuring model functions

This is the optional step referenced in both setup procedures above. The following functions change Chat Agent behavior for the model they're applied to. Apply or update them in Asset Library > Models by editing the model's Functions field:

Set chatbot-default-thinking-level only on models that support thinking or reasoning modes. If you don't set it, the model uses its own default thinking level (typically medium, per the model's specification). That default works well for most tasks. Leave the function unset unless you see inconsistent Chat Agent behavior on complex tasks or you need faster responses. Raise it to high for more complex tasks, or lower it to low to speed up responses.

Set chatbot-model-tier only when your model isn't on the recognized list. Chat Agent auto-detects strength for recognized models. See Assigning a tier to an unrecognized model for the procedure.

Verifying your configuration

1. In the HM console, sign in as a user that belongs to at least one project.
2. Select the sparkle icon at the top right.
3. Confirm the panel opens with the header HM AI and the placeholder Ask me anything... in the input.
4. If you registered more than one model, select the model picker and confirm your model appears.
5. Send a quick prompt, such as: What can you do?
6. Confirm a streamed response begins within a few seconds.

If the sparkle icon is missing or the panel can't be opened, see Troubleshooting.

Accessing the Chat Agent

Chat Agent is available on every authenticated page in the HM console.

1. Look for the rainbow sparkle icon at the top right of the HM header.
2. Select the icon to open Chat Agent.

If the icon is missing or disabled, see Troubleshooting.

Chat Agent opens in two layouts:

• Full screen — the default. The conversations list appears on the left.
• Floating panel — a compact, draggable panel docked at the bottom right of the HM console. Use the expand or shrink button to switch between modes.

Using the Chat Agent

Ask a question

1. Select Model from the picker if your administrator registered more than one. The model determines which capabilities are available — see Model-dependent capabilities.
2. Type your question in the Ask me anything input and press Enter.
3. Use Shift+Enter to add a line break within your message.
4. Chat Agent streams a response in the chat window. Documentation answers include inline citations so you can open the source page directly.
5. Select the stop button to interrupt generation early.

If Chat Agent needs more information before completing your request — for example, which project to target or a connection password — it pauses and presents a structured form. Fill in the form to continue, or cancel it (or type an unrelated message) to dismiss it.

Use page context

When you open Chat Agent from a cluster, project, or other resource page, it picks up that page's context automatically. The project and cluster IDs appear as small tags above the input, so you can ask about that resource without naming it — for example, Show me the active alerts applies to the cluster you're currently viewing.

Manage conversations

Each conversation is saved to your HM organization so you can return to it later. Select New Chat to start a fresh conversation, or open the sidebar with the menu icon to return to an earlier one.

Chat Agent gives each new conversation a short auto-generated title. From the sidebar you can also Rename a conversation or Delete one — deletion can't be undone.

For conversation privacy details, see Conversation storage and isolation.

Recommendations

• Start with a Large model — Large is the lowest tier that covers Postgres Cluster lifecycle management with structured forms and the full set of Medium-tier capabilities. gpt-4.1, claude-haiku-4-5, gemini-2.5-pro, and gpt-5 sit in this tier.
• Pick a self-hosted model if your prompts contain regulated or proprietary data — this choice keeps Chat Agent prompts and responses inside your infrastructure.
• Register more than one model to let users pick a tier per conversation — for example, register a Small model for documentation lookups and an Extra Large model for cluster-creation flows.

Supported model providers

Chat Agent communicates with models through three API protocols. Any provider — third-party or self-hosted — that speaks one of these protocols works. EDB does not build or ship AI models. All HM-hosted models are third-party models downloaded by HM from Hugging Face or NVIDIA, such as gpt-oss-20b and nvidia/llama-3.3-nemotron-super-49b-v1.5.

Model strength tiers

Chat Agent classifies every registered model into a strength tier — Small, Medium, Large, or Extra Large — based on the model's capability. A model's tier determines which Chat Agent skills are available when that model is selected.

Strength tiers

The four tiers correspond to approximate model scale and capability:

• Small — roughly 7–30B parameter class.
• Medium — roughly 30–100B parameters.
• Large — gpt-4.1 / claude-haiku-4-5 / gemini-2.5-pro class.
• Extra Large — frontier models in the gpt-5.4+, gpt-5.3-codex, claude-sonnet-4-5+, gemini-3+, and Claude Opus families.

A model that isn't on the recognized list below defaults to Small.

Recognized models

The following aliases are recognized as the same canonical model:

• Anthropic models on Amazon Bedrock — cross-region prefixes us., eu., apac., global. and provider-prefixed variants (anthropic.<model>-<date>-v1:0).
• OpenAI gpt-oss-20b and gpt-oss-120b — provider-prefixed variants openai/<model>, openai.<model>-1:0, gpt-oss:<size>.

Assigning a tier to an unrecognized model

Models whose names don't match any row in the table above (and aren't covered by the alias rules) default to Small. Many of the more advanced skills are unavailable as a result.

To override this setting, add the chatbot-model-tier function to the model in the Asset Library with one of these values:

Use the value that best matches your model's actual capability. After saving, Chat Agent honors the tier you set instead of defaulting to Small.

Skills available at each tier

A skill is available when the selected model's tier is at or above the skill's minimum tier. Higher tiers add capabilities without removing the existing ones.

At Large and Extra Large tiers, Chat Agent asks structured follow-up questions to confirm parameters before it submits Postgres cluster lifecycle management tasks. Small-tier models don't offer Postgres Cluster lifecycle management. For a full breakdown of what each tier supports, see What the Chat Agent can do.

Disabling Chat Agent for an organization

To remove Chat Agent from your HM console, remove the chatbot-gen-content function from every model that has it. Once no Chat Agent provider is registered, the sparkle icon is shown but disabled, with a tooltip directing users to configure a model service.

To remove the sparkle icon entirely, disable the AI scenario at install time. See HM installation scenarios.

Next steps

• What the Chat Agent can do — the capability catalog users see, organized by area.
• Security and data handling — the authorization model, audit, and data treatment.
• Troubleshooting — common configuration issues and how to resolve them.