Setting up the Virtana AI Copilot Chatbot

Virtana AI adds a Copilot Chatbot to your infrastructure monitoring environment so you can work with alerts, documentation, and insights through a conversational interface. The Copilot connects to an external large language model (LLM) provider that powers its responses. Before you enable the Copilot, you configure one supported provider, add the connection details to your Global View deployment, and turn on the related feature flags.

When the Copilot is enabled, you can do the following:

Interact with product documentation. Ask questions in natural language and get contextual answers drawn from Virtana's documentation.
Query alert-related information. Retrieve active alerts, alert history, and alert configurations without writing a query.
Generate alert policies and insights. Create alert policies and surface actionable insights based on data from your environment.

Supported model providers

This section describes the large language model (LLM) providers that Virtana AI supports. You must configure exactly one provider before you enable the Copilot, because the provider supplies the model that powers the chatbot, alert generation, and insights features.

The following table describes the three supported providers and when to use each one:

Provider	When to use it
OpenAI	Choose OpenAI to connect directly to the OpenAI API and its hosted models (for example, GPT-4o and GPT-5).
Gemini	Choose Gemini to connect to Google's Gemini models.
LiteLLM	Choose LiteLLM when you want a single gateway in front of multiple model backends, or when you run self-hosted or custom model deployments. LiteLLM provides a unified API across those backends.

LiteLLM model configuration

This section applies only if you select LiteLLM as your provider. In your LiteLLM instance, you create one model for each Virtana AI function and give each model the exact name shown in the following table. All of the model names are required, because each name maps to a specific function and Virtana AI calls them by name.

VIRTANA_MDL_MINI_DEFAULT_CONFIG: 'gpt-4o-mini'
VIRTANA_MDL_MINI_CHATOPENAI: 'gpt-4o'
VIRTANA_MDL_MINI_ROUTEOPENAI: 'gpt-4o-mini'
VIRTANA_MDL_MINI_SWAGGER_ANALYSIS: 'gpt-4o-mini'
VIRTANA_MDL_REASON_INSIGHT_SVC: 'gpt-4o'
VIRTANA_MDL_MINI_GV_ALERT: 'gpt-4o-mini'
VIRTANA_MDL_MINI_GV_ALERT_TOOL_SELECTOR: 'gpt-4o-mini'
VIRTANA_MDL_REASON_GV_ALERT_RESPONSE_GENERATOR: 'gpt-4o'
VIRTANA_MDL_MINI_CHART_GENERATOR: 'gpt-4o-mini'
VIRTANA_MDL_REASON_METRICS: 'gpt-4o'
VIRTANA_MDL_MINI_METRICS_STATEMENT: 'gpt-4o-mini'
VIRTANA_MDL_REASON_TROUBLESHOOTING: 'gpt-4.1'
VIRTANA_MDL_NANO_IDENTIFIER: 'gpt-4.1-nano'
VIRTANA_MDL_MINI_GENERAL: 'gpt-4o-mini'
VIRTANA_MDL_REASON: 'gpt-5'
VIRTANA_MDL_REASONING_MINI: 'gpt-5-mini'

The following table lists the required model name, the model it maps to, and the Virtana AI function it serves:

Model name	Mapped model	Virtana AI function
`VIRTANA_MDL_MINI_DEFAULT_CONFIG`	`gpt-4o-mini`	Provides the default configuration for lightweight tasks.
`VIRTANA_MDL_MINI_CHATOPENAI`	`gpt-4o`	Powers the primary chatbot conversations.
`VIRTANA_MDL_MINI_ROUTEOPENAI`	`gpt-4o-mini`	Routes each user query to the correct handler.
`VIRTANA_MDL_MINI_SWAGGER_ANALYSIS`	`gpt-4o-mini`	Analyzes API specifications in Swagger or OpenAPI format.
`VIRTANA_MDL_REASON_INSIGHT_SVC`	`gpt-4o`	Generates detailed insights from monitoring data.
`VIRTANA_MDL_MINI_GV_ALERT`	`gpt-4o-mini`	Handles GlobalView alert queries.
`VIRTANA_MDL_MINI_GV_ALERT_TOOL_SELECTOR`	`gpt-4o-mini`	Selects the correct tool for an alert-related task.
`VIRTANA_MDL_REASON_GV_ALERT_RESPONSE_GENERATOR`	`gpt-4o`	Generates detailed responses for alert queries.
`VIRTANA_MDL_MINI_CHART_GENERATOR`	`gpt-4o-mini`	Creates chart visualizations from data.
`VIRTANA_MDL_REASON_METRICS`	`gpt-4o`	Processes and reasons over metrics data.
`VIRTANA_MDL_MINI_METRICS_STATEMENT`	`gpt-4o-mini`	Generates natural-language summaries of metrics.
`VIRTANA_MDL_REASON_TROUBLESHOOTING`	`gpt-4.1`	Powers advanced troubleshooting workflows.
`VIRTANA_MDL_NANO_IDENTIFIER`	`gpt-4.1-nano`	Performs lightweight entity identification and classification.
`VIRTANA_MDL_MINI_GENERAL`	`gpt-4o-mini`	Handles general-purpose lightweight tasks.
`VIRTANA_MDL_REASON`	`gpt-5`	Serves as the primary reasoning model for complex analysis.
`VIRTANA_MDL_REASONING_MINI`	`gpt-5-mini`	Provides lightweight reasoning for simpler analytical tasks.

Note

The GPT models mapped in the table above are the officially tested and validated models for the Virtana AI chatbot, alert policy generation, and insights generation. If you map names to models other than those listed, Virtana AI might behave inconsistently or in ways that aren't supported.

After you create all of the models in LiteLLM, generate an API key that has access to every model name in the table above. You add this key to the Global View configuration.

Global View configuration

This section describes how to connect Global View to your chosen LLM provider and turn on Virtana AI. You add a configuration block to your global-view-values.yaml file that enables the feature, names the provider, and supplies the credentials. The same block applies whether you use OpenAI, Gemini, or LiteLLM — only the values you enter differ.

Add the following block to global-view-values.yaml:

virtana-ai:
  enabled: true
  env:
    VIRTANA_AI_MODEL_PROVIDER: ""  

cp-configs:
  litellm:
    litellm_api_key: ""
    litellm_inference_endpoint: ""

Parameter descriptions

The following table describes each parameter in the configuration block, including the value to enter and when the value is required:

Parameter	Description and value
`virtana-ai.enabled`	Turns the Virtana AI Copilot Chatbot feature on or off. Set this to `true` to enable the Copilot.
`virtana-ai.env.VIRTANA_AI_MODEL_PROVIDER`	Names the LLM provider that powers Virtana AI. Enter one of the supported values: `OPENAI`, `GEMINI`, or `LITELLM`. The value must match the provider you configured.
`cp-configs.litellm.litellm_api_key`	Authenticates Virtana AI with your provider. Enter the API key for your chosen provider. This key is required for all three providers (OpenAI, Gemini, and LiteLLM). For LiteLLM, use the key that has access to every model name in the LiteLLM model configuration.
`cp-configs.litellm.litellm_inference_endpoint`	Sets the inference endpoint URL that Virtana AI calls. Leave this blank for OpenAI and Gemini, because they use their default public endpoints. If you run a self-hosted LiteLLM instance, enter the full endpoint URL, for example, `https://litellm.example.com`.

Enable GrowthBook feature flags

This section describes the feature flags that activate the Copilot and insights features for your end users. GrowthBook is the feature-flag service that controls which features appear in the product. After you deploy Global View with the configuration described in Global View configuration, set the two flags below in GrowthBook.

The following table lists each feature flag, the value to set, and the effect of that value:

Feature flag	Default value	Effect
`vp_disable_copilot`	`false`	Enables the Copilot Chatbot in the user interface. The flag controls the disable behavior, so a value of `false` turns the Copilot on.
`ipm_virtana_ai_insights`	`true`	Enables AI-powered insights generation in the IPM module.

In this section: