OpenAI Releases Realtime Audio Models 2.1 and Mini

pbbakkum · x · 2026-07-07

OpenAI has released two new audio models. The gpt-realtime-2.1 is an incremental update to realtime-2, significantly improving the handling of alphanumeric details like phone numbers and emails during conversations. Meanwhile, the gpt-realtime-2.1-mini is a smaller version that still retains reasoning capabilities.

Related event: OpenAI Launches GPT-Realtime-2.1 Audio Models(4 posts)→

Original post →

More from Multimodal

Multimodal channel →