### Answer I don't have specific documentation explaining why token usage would increase by ~2000 tokens without any changes to your agent configuration. However, here are some potential factors that could affect token usage: 1. Knowledge Base retrieval — if your agent uses a knowledge base, tokens are consumed when content is retrieved and injected into the conversation context [(1) ]() 2. Conversation length — longer conversations accumulate more tokens as context is maintained throughout the call [(2) ]() 3. Platform updates — Retell has made various updates including changes to LLM prompt token limits (now set to 32,768 tokens) [(3) ]() To investigate further, I'd recommend: - Checking if any platform updates affected your agent - Reviewing the

llm_token_usage

field in your call data via the Get Call API to see token breakdown [(4) ]() For a definitive answer on unexpected token increases, I'd suggest reaching out to Retell support directly: - [Contact Sales]() - [Retell AI GitHub]() Would you like me to try providing a general answer based on how LLM token usage typically works?

​ To learn more, see the sources I considered: [[1] Create Web Call](https://docs.retellai.com/api-references/create-web-call) [[2] Choosing the Best LLM for Your AI Voice Agents | Retell AI](https://www.retellai.com/blog/choosing-the-best-llm-for-your-voice-ai-agents) [[3] Platform Changelogs | Retell AI](https://www.retellai.com/changelog) [[4] Get Call](https://docs.retellai.com/api-references/get-call) ​

Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.