impulse

GPT Realtime

Frontier labRealtimeRetiring 20 Jan 2027
API id gpt-realtimeReleased 28 Aug 2025 (Aug 2025)
OpenAI retires it on 20 Jan 2027 (in 63 days); Azure OpenAI on 2 Mar 2027. No API serves it after that.
Input / 1M
$4.00
at OpenAI
Output / 1M
$16.00
at OpenAI
Context window
32K
tokens
Max output
4K
tokens per reply
Hosts
2
4 listings
Released
Aug 2025
28 Aug 2025

Pricing

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

TierInput / 1MOutput / 1M
Standard$4.00$16.00
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$2,076
  • RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$1,632
  • Coding agent1,000 requests a day, 40K in / 2K out, cached$2,736
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$13,440
Use your own numbers

Where to run it

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

HostIn / out per 1M
OpenAILab API
gpt-realtime
$4.00 / $16.00
Retires 20 Jan 2027
gpt-realtime
$4.00 / $16.00
Retires 2 Mar 2027
Show 2 more listings: regions, snapshots and variants
OpenAI
2025-08-28
gpt-realtime-2025-08-28
$4.00 / $16.00
Retires 20 Jan 2027
Azure OpenAI
2025-08-28
gpt-realtime-2025-08-28
$4.00 / $16.00
Retires 2 Mar 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.

Call it

Using OpenAI's API.

Python
from openai import OpenAI

client = OpenAI()  # reads OPENAI_API_KEY

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

Quick answers

How much does GPT Realtime cost?

On OpenAI, GPT Realtime costs $4.00 per million input tokens and $16.00 per million output tokens. Cached input costs $0.40 per million tokens.

Which APIs offer GPT Realtime?

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

What is GPT Realtime's context window?

32K tokens, with up to 4K tokens in one reply.

Is GPT Realtime being retired?

OpenAI retires it on 20 Jan 2027; Azure OpenAI on 2 Mar 2027. No API serves it after that.

What are cheaper alternatives to GPT Realtime?

Models that cost less and keep most of what it can do: GPT Realtime 2.1 Mini on OpenAI (85% less), Gemini 3.8 Live on Google AI Studio (76% less) and Gemini 3.8 Live Extended Thinking on Google AI Studio (76% less). Savings are on a mix of 3 input tokens to 1 output token.

When was GPT Realtime released?

OpenAI released it on 28 Aug 2025.

What is GPT Realtime's API model id?

Use gpt-realtime with OpenAI.

Sources

Raw catalog entry for gpt-realtime 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,
  "deprecation_date": "2027-01-20",
  "deprecation_source": "https://developers.openai.com/api/docs/deprecations",
  "successor_model": "gpt-realtime-2.1",
  "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": 32000,
  "max_output_tokens": 4096,
  "mode": "realtime",
  "output_cost_per_audio_token": 0.000064,
  "output_cost_per_token": 0.000016,
  "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
}