GPT Realtime

Legacy OpenAI speech-to-speech realtime model, superseded by gpt-realtime-2.1. Served via the gateway's /v1/realtime WebSocket endpoint with text and audio input/output and function calling.

gpt-realtime
BETAScheduled for DeactivationGet StartedView uptime
32,768 context
Released August 28, 2025
Starting at $4.00/M input tokens
Starting at $16.00/M output tokens
Tools
No ratings yetSign in to rate

Select Provider

All Providers for GPT Realtime

LLM Gateway routes requests to the best providers that are able to handle your prompt size and parameters.

OpenAI
Context: 32.8k
Deprecated since Jul 20, 2026Deactivating on Jan 20, 2027
Input
$4
/M tokens
Cache Read
$0.4
/M tokens
Output
$16
/M tokens
Audio Output
$64
/M tokens
Get Started

Frequently asked questions

What is GPT Realtime?

Legacy OpenAI speech-to-speech realtime model, superseded by gpt-realtime-2.1. Served via the gateway's /v1/realtime WebSocket endpoint with text and audio input/output and function calling. You can access it through LLM Gateway's OpenAI-compatible API with automatic provider routing, fallback, and cost analytics.

How much does GPT Realtime cost?

Pricing for GPT Realtime on LLM Gateway starts at $4.00 per million input tokens and $16.00 per million output tokens, depending on the provider. The pricing table above always reflects the current per-provider rates.

What is the context length of GPT Realtime?

GPT Realtime supports a context window of up to 32,768 tokens on its largest provider deployment.

Which providers serve GPT Realtime?

GPT Realtime is served by OpenAI through LLM Gateway. Requests are automatically routed to the best available provider, with fallback when a provider has issues.

Does GPT Realtime support tool calling and structured outputs?

GPT Realtime supports tool (function) calling, but not structured JSON output mode.

When was GPT Realtime released?

GPT Realtime was released on August 28, 2025.