### Answer Based on the available sources, I can confirm that GPT-5.5 and GPT-5.4 models are among the available LLM options in Retell AI [(1) ]() [(2) ]() . The available models include gpt-5, gpt-5-mini, gpt-5-nano, gpt-5.1, gpt-5.2, gpt-5.4, gpt-5.4-mini, gpt-5.4-nano, and gpt-5.5 . However, I don't have specific information in the sources about why GPT-5.5 and GPT-5.4 are now the top suggested LLMs or what specific benefits they offer compared to earlier recommended models like GPT-4.1. What I can tell you about LLM selection in general: Temperature settings can significantly impact your agent's behavior [(3) ]() . Lower temperatures (0.0-0.3) provide highly consistent, deterministic responses best for function calling and data collection, while higher temperatures (0.4-0.7) offer balanced consistency and variation suitable for general customer service . The sources indicate that some older models have been deprecated. For instance, Claude 3.7 Sonnet was deprecated by Claude, and agents using deprecated models are moved to the GPT-4.1 model [(4) ]() . Additionally, there have been reports that GPT-5 had hallucination issues when it was new, with recommendations to stick with GPT-4.1 until it stabilized [(5) ]() . For specific details about the benefits of GPT-5.5 and GPT-5.4 compared to earlier models, I'd recommend checking the Retell AI dashboard or reaching out to support for the most current information about model performance and recommendations.

​ To learn more, see the sources I considered: [[1] Update Retell LLM](https://docs.retellai.com/api-references/update-retell-llm) [[2] Create Retell LLM](https://docs.retellai.com/api-references/create-retell-llm) [[3] Configure LLM Options](https://docs.retellai.com/build/llm-options) [[4] Some LLMs are missing from the selection list now ?](https://discord.com/channels/1186424376300679209/1437391224171659307) [[5] GPT 5 Allucination](https://discord.com/channels/1186424376300679209/1403421375288447117) ​

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

Appreciate the feedback! Feel free to ask me again with additional guidance or context.