Hi! Will GPT-5-Chat-latest be available soon, please? Thanks ! 🙂

Thank you for the request. We will look into it and get back to you. Would you be open to sharing why you use gpt 5 chat? How’s the latency and the conversation quality and tool calling?

Hello ! 🙂 The prompt is better than 4o on our end; we have a large, very precise prompt. The latency is generally the same as 4o as well. And the tools work better than with 4o because it understands the context of the prompt better. The answers are much more natural than 4o: We work with Alzheimer's patients using voice cloning of their loved ones, so in terms of testing, we can't do better! Do you think we could have it soon? The models are hosted on your servers; are there any in Europe? Thank you very much! 👍

Thank you very much for sharing about your experiences with gpt-5-chat-latest. I've heard from another user that it is also as fast as 4o, slightly cheaper, and smart. We are still working on our plan internally on which new models to support in LiveKit Inference. If we end up including support for 5-chat, I will let you know. LiveKit Inference is not GDPR compliant yet. Users can't specify that their model requests and responses stay in a particular region, Scale tier customers can request region pinning for media transport, but not inference.

Great, just let me know when the model is available! Thanks! 👍 cc @bitter-appointment-81509