I'm looking to get the lowest latency possible. This is my current latency, model is my custom llm. https://cdn.discordapp.com/attachments/1403078982420795592/1403079315460980756/Screenshot_2025-08-07_at_11.15.58_AM.png?ex=68963ef0&is=6894ed70&hm=7941cefdbbca596a0a153108c53a3ee7c48f2333fe7bbcb9e39841a1ee74fba7&

To achieve the lowest latency for your Custom LLM completions endpoint, deploy it in a region geographically closest to your primary users or to the Vapi infrastructure. For example, if most users are in the US East, choose a US East data center. > Tip: Regional deployment is key for reducing latency and ensuring fast response times ([DeepInfra documentation](https://docs.vapi.ai/providers/model/deepinfra)). Source: - [DeepInfra documentation](https://docs.vapi.ai/providers/model/deepinfra)

I'm currently on us-east-1

Hi, Azure is a good choice for US east due to the region selection. They have a US east server