Configure Agent Model in Geta.ai

Steps to Configure Agent Model

1) Navigate to Agent Model

• Log in to your Geta.ai dashboard.

• Open the Chatbot builder then Voice Bot module.

• Select the specific voice bot you’ve created.

• Click on the Agent Setup.

2) Set Temperature (Creativity Level)

• Use the Temperature slider to control how creative or precise your Voice Agent’s responses should be.

  • Lower value (e.g., 0.2–0.4): Focused and consistent replies.

  • Medium value (e.g., 0.5–0.7): Balanced and conversational tone.

  • Higher value (e.g., 0.8–1.0): More creative and varied responses.

💡 Tip: For academic or factual bots, keep the temperature lower for accuracy.

3) Choose the LLM Model

• Under LLM Model, select which AI language model you want your bot to use.

• Available options include:

  • GPT-4o, GPT-4o Mini, GPT-4 Turbo

  • Gemini 1.5 Pro, Gemini 2.0 Flash

  • Claude 3.7 Sonnet, Claude 3.5 Haiku, Grok Beta, and more.

Each model offers different speed, accuracy, and creativity levels.
Choose based on your use case — for instance:

• GPT & Claude models → High reasoning and accuracy.

• Gemini Flash or GPT Mini models → Faster, lightweight performance.

4) Set Limit Token Usage

• The Limit Token Usage defines the maximum text length the model can process or generate in a single response.

• Example: A limit of 1000 tokens equals roughly 750 words.

• Lower values → shorter, faster replies.

• Higher values → detailed answers but more processing time.

💡 Recommended: Keep it between 800–1200 for voice interactions.

5) Upload Knowledge Base

• In the Knowledge Base section, upload all relevant files that your Voice Agent should refer to while responding to queries.

• Click on Upload File and add PDFs such as:

  • Course brochures, admission documents, fee structures, etc.

• You can upload multiple files — the agent will scan and reference these files to give accurate, context-based answers.

💡 Ensure your PDFs are text-based, not scanned images, for best accuracy.

Save Configuration

• After setting Temperature, Model, Token Limit, and uploading files, click Save Changes.

• This will apply your Agent Model configuration to the selected Voice Bot.

Next Steps

• Set the Voice: Choose your agent’s voice tone, gender, and speaking pace under the Voice tab.

• Analyze Performance: Review the bot’s performance and analytics once the setup is complete.

• Advance Settings: Fine-tune advanced configurations to enhance control and response behavior.