Skip to main content

​
What is the Intelligence tab?

The Intelligence tab (previously labeled LLM) is where you select and configure the intelligence behind your voice AI agent. Choose your language model provider, adjust response parameters, and connect knowledge bases for enhanced conversations.
Intelligence Tab in Agent Studio showing Choose LLM model with openai and gpt-4.1-mini, Model Parameters with token and temperature sliders, and a Knowledge Base multi-select

Intelligence Tab with the Choose LLM model, Model Parameters and Knowledge Base cards

​
Configuration options

​
Choose LLM model

Select your AI provider and model for conversation intelligence.
Choose LLM model card with Provider set to openai and Model set to gpt-4.1-mini

Choose LLM model card with the Provider and Model dropdowns

Provider selection

Choose from Azure, OpenAI, Anthropic, Groq, and more

Model selection

Pick the specific model (e.g., gpt-4.1-mini)
Connect your own provider keys in Settings → Integrations to reduce costs and access more models.

​
Model parameters

Fine-tune how your agent generates responses.
Model Parameters card with Tokens generated on each LLM output set to 370 and Temperature set to 0.35, each with a slider and numeric input

Model Parameters card with the token limit and temperature sliders

Keep temperature low (0.3-0.5) if you want consistent, controlled responses. Higher temperature increases creativity but may cause deviation from your prompt instructions.GPT-5 models are the exception: they accept only 1, so the temperature control has no effect on them. Constrain those agents through the prompt instead.
On GPT-5 models, reasoning tokens are drawn from the same budget as Tokens generated on each LLM output, so raise the cap whenever you raise reasoning effort. The effort control lists only the values the selected model accepts, and the options change when you switch models. See OpenAI for the per-model values.

​
Extra request params

With a custom model selected, the tab also shows Extra request params. Switch it on to send your server parameters beyond the OpenAI standard as JSON, for example {"chat_template_kwargs": {"enable_thinking": false}} to turn Qwen3’s thinking off on vLLM. The example buttons fill in common ones. See Extra request params.
Model Parameters card with the Extra request params toggle switched on, a JSON box containing chat_template_kwargs with enable_thinking set to false, and example buttons for Thinking off (vLLM / SGLang) and Reasoning off (OpenRouter)

Extra request params under Model Parameters, shown when a custom model is selected

​
Add knowledge base

Connect your knowledge bases to give your agent accurate, contextual information.
Knowledge Base card with the Select knowledge bases dropdown open, listing a PDF and a URL knowledge base with checkboxes and an Add new knowledgebase link

Knowledge Base card with the multi-select dropdown open

1

Click the Dropdown

Open the “Select knowledge bases” multi-select dropdown.
2

Select Knowledge Bases

Check one or more knowledge bases (PDFs, URLs) to connect.
3

Create New (Optional)

Click “Add new knowledgebase” to create and upload new content.
Knowledge bases enable your agent to answer questions with accurate, up-to-date information from your documents and URLs. Connect multiple knowledge bases for comprehensive coverage.
Create knowledge bases in the Knowledge Base section by uploading PDFs or adding URLs.

​
Next steps

Knowledge base

Create and manage knowledge bases

Agent tab

Configure prompts and welcome message

Languages tab

Set up voice, transcription, and languages

LLM providers

Connect your own LLM provider