impulse

GPT Realtime 2

Frontier labRealtime
API id gpt-realtime-2
A newer model in this line is out: GPT Realtime 2.1, released 7 Jul 2026.
Azure OpenAI retires it on 20 Jan 2027 (in 63 days). 1 host keeps serving it.
Input / 1M
$4.00
at OpenAI
Output / 1M
$24.00
at OpenAI
Context window
128K
tokens
Max output
32K
tokens per reply
Hosts
2
2 listings
Released
-
date not published

Pricing

List prices at OpenAI, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$4.00$24.00
Cached input
90% off input
$0.40

What it costs a month

At OpenAI list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$2,676
  • RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$1,920
  • Coding agent1,000 requests a day, 40K in / 2K out, cached$3,216
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$15,360
Use your own numbers

Where to run it

2 hosts serve GPT Realtime 2. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
OpenAILab API
gpt-realtime-2
$4.00 / $24.00
gpt-realtime-2
$4.00 / $24.00
Retires 20 Jan 2027

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
AudioImagesText
Produces
AudioText
Features
Audio inputAudio outputParallel tool calls

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

No host sells it for less than OpenAI.

Smaller OpenAI models

Other frontier labs

Open-weight picks

No cheaper open-weight match.

Model line

Newer and older versions, at each lab's own price.

  1. Jul 2026GPT Realtime 2.1Latest$4.00 / $24.00
  2. UndatedGPT Realtime 2This page$4.00 / $24.00
  3. Feb 2026GPT Realtime 1.5$4.00 / $16.00

Call it

Using OpenAI's API.

Python
from openai import OpenAI

client = OpenAI()  # reads OPENAI_API_KEY

response = client.responses.create(model="gpt-realtime-2", input="Hello")
print(response.output_text)

Quick answers

How much does GPT Realtime 2 cost?

On OpenAI, GPT Realtime 2 costs $4.00 per million input tokens and $24.00 per million output tokens. Cached input costs $0.40 per million tokens.

Which APIs offer GPT Realtime 2?

GPT Realtime 2 runs on 2 APIs: OpenAI and Azure OpenAI.

What is GPT Realtime 2's context window?

128K tokens, with up to 32K tokens in one reply.

Is GPT Realtime 2 being retired?

Azure OpenAI retires it on 20 Jan 2027. 1 host keeps serving it.

Is there a newer version of GPT Realtime 2?

Yes. GPT Realtime 2.1 is the newest model in this line, released on 7 Jul 2026.

What are cheaper alternatives to GPT Realtime 2?

Models that cost less and keep most of what it can do: GPT Realtime 2.1 Mini on OpenAI (88% less), Gemini 3.8 Live on Google AI Studio (81% less) and Gemini 3.8 Live Extended Thinking on Google AI Studio (81% less). Savings are on a mix of 3 input tokens to 1 output token.

What is GPT Realtime 2's API model id?

Use gpt-realtime-2 with OpenAI.

Sources

Raw catalog entry for gpt-realtime-2 on OpenAI
{
  "cache_creation_input_audio_token_cost": 4e-7,
  "cache_read_input_audio_token_cost": 4e-7,
  "cache_read_input_image_token_cost": 5e-7,
  "cache_read_input_token_cost": 4e-7,
  "input_cost_per_audio_token": 0.000032,
  "input_cost_per_image_token": 0.000005,
  "input_cost_per_token": 0.000004,
  "litellm_provider": "openai",
  "max_input_tokens": 128000,
  "max_output_tokens": 32000,
  "mode": "realtime",
  "provider_specific_entry": {
    "realtime_ga_only": 1
  },
  "output_cost_per_audio_token": 0.000064,
  "output_cost_per_token": 0.000024,
  "source": "https://developers.openai.com/api/docs/pricing",
  "supported_endpoints": [
    "/v1/realtime"
  ],
  "supported_modalities": [
    "text",
    "image",
    "audio"
  ],
  "supported_output_modalities": [
    "text",
    "audio"
  ],
  "supports_audio_input": true,
  "supports_audio_output": true,
  "supports_function_calling": true,
  "supports_parallel_function_calling": true,
  "supports_system_messages": true,
  "supports_tool_choice": true
}