← Back to all models
gpt-realtime-2.1
OpenAI · gpt-realtime-2.1
GPT Realtime 2.1 — گفتار به گفتار با تأخیر کم (Azure)
Starting price
1,016,400
per million tokens
Billing
Pay as you go
Service status
Active
About gpt-realtime-2.1
GPT Realtime 2.1 — گفتار به گفتار با تأخیر کم (Azure)
gpt-realtime-2.1 pricing
Input
1,016,400
Output
6,098,400
Cache (read)
1,016,400
same as input
Cache (write)
1,016,400
same as input
Model id
Provider
Context window
Billing basis
Status
Share model
Sample code and API for gpt-realtime-2.1
Point the base URL at CupAI and use your own API key; the rest of the request stays exactly as it is in any compatible client.
# npm i -g wscat
wscat -c 'wss://api.cupai.ir/v1/realtime?model=gpt-realtime-2.1' \
-H 'Authorization: Bearer $CUPAI_API_KEY'
# then send, in order:
# {"type":"session.update","session":{"type":"realtime"}}
# {"type":"response.create"}Replace CUPAI_API_KEY with your own key.
Create an API key
Frequently asked questions about gpt-realtime-2.1
- How is usage of gpt-realtime-2.1 billed?
- By the input and output tokens of each request. The rate for each part is in the pricing table on this page, and the amount is deducted from your wallet.
- How do I connect to gpt-realtime-2.1?
- Point the base URL at CupAI, put your API key in the request header, and send the model id in the request body. Ready-made samples are on this page.
- How do payments work?
- Payments are in Toman and need no foreign credit card. Credit you buy can be spent on any available model.
Other models
How model pricing works
Every AI model strikes a different balance between speed, answer quality and cost. Flagship models suit complex reasoning, long-document analysis and precise code generation, while lighter models cost considerably less for summarising, classification and short answers. The list above shows each model's rate broken down by input token, output token and cache, so you can estimate the cost of your workload before you start.
A token is the smallest unit of text a model processes; roughly speaking a thousand tokens is about 750 English words. The cost of a request is the sum of its input and output tokens, so shortening your prompt and capping the response length reduces cost directly. Some models have a second price tier for very long inputs, which reprices the entire request once it crosses a stated threshold.