GPT Realtime
OpenAIGPT Realtime is OpenAI's model for live voice sessions — sound, text and pictures go in, speech and text come back, over a WebSocket held open while the conversation lasts. It carries a 32k context with vision and tool calling, and answers only on the realtime endpoint, never on chat completions.
The real OpenAI model
Same weights, same API, same answers. We change only how you pay for it: one key for every maker, one bill, no separate account to open. Every figure on this page comes from OpenAI’s own page — check it.
Maker's own pagePrice
input / output · per 1M tokensOpenAI
Official price
Sound
$32.00 / $64.00
Text is priced above; sound is priced here, in and out. OpenAI bills the two apart and so do we.
Context
32,000
up to 4,096 out
4,096 by default here
Modalities
What it would cost you
1000 tokens is roughly 750 English words.
How it is served
Requests are retried automatically if a provider is unavailable.
Similar models
Models built for the same kind of work, at a similar price.
We do not mark up the price of tokens. It is the same price the maker charges.
Use it from your code
from openai import OpenAI
# Works with the official openai package (pip install openai)
client = OpenAI(
base_url="https://api.flintbeam.com/v1",
api_key="sk-live-your-api-key"
)
response = client.chat.completions.create(
model="openai/gpt-realtime",
messages=[
{"role": "system", "content": "You are an experienced software engineer."},
{"role": "user", "content": "Describe the architecture of a distributed cache."}
],
temperature=0.7,
max_tokens=1500
)
print(response.choices[0].message.content)