LLM Reference
OpenAI API

Using GPT Realtime 2 on OpenAI API

Implementation guide · GPT Realtime 2 · OpenAI

Serverless

OpenAI API exposes GPT Realtime 2 through model ID gpt-realtime-2. Use the setup steps, sourced pricing, capabilities, and official provider links below to validate this route before deployment.

Last refreshed 2026-05-16. Next refresh: weekly.

Quick Start

  1. 1
    Create an account at OpenAI API and generate an API key.
  2. 2
    Use the OpenAI API SDK or REST API to call gpt-realtime-2 — see the documentation for request format.
  3. 3
    You'll be billed $32.00/1M input, $64.00/1M output tokens. See full pricing.

Code Examples

Install
pip install openai
API key
OPENAI_API_KEY
Model ID
gpt-realtime-2

Use the model slug directly, e.g. "gpt-4o" or "gpt-4.1-mini".

import os
from openai import OpenAI

client = OpenAI()  # reads OPENAI_API_KEY from env
response = client.chat.completions.create(
    model="gpt-realtime-2",
    messages=[{"role": "user", "content": "Hello"}]
)
print(response.choices[0].message.content)

Pricing on OpenAI API

TypePrice (per 1M)
Input tokens$32.00
Output tokens$64.00

Capabilities

MultimodalReasoningJSON / Tool usePrompt CachingAudio

About GPT Realtime 2

GPT Realtime 2 is OpenAI's second-generation real-time voice model, released May 7, 2026. It is a GPT-5-class speech-to-speech model for voice agents with five reasoning intensity levels, parallel tool calls, spoken preambles, and recovery behavior on failed tasks. The model supports audio and text interaction through the Realtime API with a 128K token context window. Audio token pricing is $32 per 1M input tokens, $0.40 per 1M cached input tokens, and $64 per 1M output tokens.

Model Specs

Released2026-05-07
Context131k
ArchitectureDecoder Only
Knowledge cutoff2024-09

Provider

OpenAI API
OpenAI API

OpenAI

San Francisco, California, United States