Skip to main content
Agents are the core of the VIVI platform. An agent is your virtual team member that interacts with customers, employees, or systems. Agents combine a number of components, such as knowledge bases and prompts, into one system. These components allow your agent to deliver helpful, accurate, and efficient support. Unlike traditional decision trees or scripted bots, VIVI agents are dynamic — they reason, act, and adapt to context. Agents can be duplicated, customized, or scaled as your organization evolves. At minimum, an agent requires a name, prompt, and model to be live. Once configured, you can use the Test Agent feature to validate responses before deployment.

​
Building Agents

When building your agent, you can customize every detail to match your organization’s standards, tone, and goals.
  1. Set a Name
    Choose a clear name that reflects the agent’s purpose.
  2. Add a Prompt
    Define the agent’s role, tone, and rules of engagement.
  3. Select a Model
    Choose the large language model that powers the agent (see Models section below).
  4. Connect Knowledge Bases
    Provide the sources of truth the agent can reference.
  5. Enable Integrations
    Connect apps, APIs, or even other agents for advanced actions.
  6. Select Channels
    Decide where the agent will be active (web chat, WhatsApp, API, etc.).
  7. Test the Agent
    Use the built-in testing feature to verify accuracy and performance before deployment.

​
Models

VIVI supports a range of large language models (LLMs), each suited to specific use cases. Below is an overview of the current model lineup:
ModelInput (tokens)Maximum Output (tokens)Use Case
gpt-4.11,000,00032,768Flagship model for complex work requiring a long context window.
gpt-4.1-mini1,000,00032,768Faster, lower-cost option for long-context tasks, with reduced accuracy.
gpt-4o128,00016,384Multimodal generalist (text, image, audio). Balanced performance.
gpt-o3200,000100,000Strong reasoning for hard problems that require step-by-step analysis.
gpt-o4-mini200,000100,000Compact reasoning model that is strong in math, coding, and visual.

​
Best Practices

  • Use descriptive names so team members can quickly identify each agent’s role.
  • Match the model to the task: choose large-context models for complex reasoning and smaller, faster models for lightweight workflows.
  • Test thoroughly after configuring each component to ensure accuracy and reliability.
⌘I