### Answer Here are some strategies to address inconsistent agent responses: Adjust LLM Temperature [(1) ]() If your agent's responses are inconsistent, try adjusting the LLM temperature . Lower temperature settings produce more consistent, deterministic outputs. Change to a More Capable Model If instructions are concise but the agent struggles to follow them consistently, try using a more capable LLM model for your agent . Split Complex Nodes For conversation flow agents, if a node contains multiple instructions and the agent inconsistently follows them, split the node into multiple focused nodes . For example, instead of one node collecting name, phone, and address, create three separate nodes . Add Conversation Finetune Examples To achieve a specific response style consistently, add conversation finetune examples . Test Your Agent [(2) ]() Use simulation testing and batch testing to identify issues early . Retell AI offers these testing methods to ensure reliability and efficiency . Since LLMs can sometimes produce inconsistent or unexpected results, running tests multiple times helps ensure more reliable and accurate outcomes . Review Transition Conditions [(1) ]() If you're using conversation flows, ensure transition conditions precisely match your intended triggers . Consider breaking down complex conditions into multiple simpler ones . If the issue persists after trying these approaches, you might want to share a specific call ID with the Retell team for deeper investigation.

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

​ To learn more, see the sources I considered: [[1] Debug guide](https://docs.retellai.com/build/conversation-flow/debug-guide) [[2] How to Deploy Conversational AI | Retell AI](https://www.retellai.com/blog/how-to-deploy-conversational-ai) ​

### Answer To lower the temperature in Conversation Flow, you have two options: Global Temperature Setting [(1) ]() Click on the empty canvas and access the global settings. Under "Select a Language Model," you can tune the LLM temperature to make answers more stable (lower temperature) or more variant (higher temperature) . Per-Node Temperature Setting [(2) ]() If you want different temperature settings for specific nodes, you can change the node model and adjust temperature for individual nodes . This is useful when certain parts of your conversation need more consistency than others. Regarding Sharing Your Flow Privately I'm an AI assistant and cannot receive private information or review your specific conversation flow. For private support with your configuration, I recommend: - Visit [Documentation]() - Visit [Hire certified Retell partner]() The Retell team and certified partners can provide personalized assistance with your specific setup.

​ To learn more, see the sources I considered: [[1] Step 1: Configure global settings](https://docs.retellai.com/build/conversation-flow/global-setting) [[2] Debug guide](https://docs.retellai.com/build/conversation-flow/debug-guide) ​

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