Very Excited @User Just integrated OpenAI new model GPT-4o with Vapi using custom llm and it is slightly better than gpt 3.5 in terms of latency https://cdn.discordapp.com/attachments/1211483571454218261/1239650022165970984/image.png?ex=6643b18c&is=6642600c&hm=96fe3ca5a8c26990f21d47a1f51424a64266576e48fbc1e6bedff17f0f2b7b85&