Mini toolsGPT Realtime 2.1
By OpenAI
Model family: GPT
GPT Realtime 2.1 updates GPT Realtime 2 with improved alphanumeric recognition, silence and noise handling, and interruption behavior for speech to speech voice agents. It processes text, audio, and image input and produces text and audio output, with a 128,000 token context window and 32,000 max output tokens. The model supports configurable reasoning effort, letting developers trade off latency against response quality, and includes function calling for connecting to external tools during live voice sessions. It is built for low latency conversational voice agents that listen, reason, speak, and call tools in real time, and is accessed through the Realtime API over WebRTC, WebSocket, or SIP.
New
Multimodal
Overview
GPT Realtime 2.1 is OpenAI's realtime speech to speech model, an update to GPT Realtime 2 with improved alphanumeric recognition, better silence and noise handling, and more natural interruption behavior. It supports configurable reasoning effort, stronger instruction following, and reliable tool use for voice agent workflows.
About OpenAI
OpenAI is a technology company that specializes in artificial intelligence research and innovation.
Industry:
Artificial Intelligence
Company Size:
4500
Location:
San Francisco, California, US
Website:
openai.com
Last updated: July 8, 2026
