GPT Realtime API Pricing
As of October 10, 2026, GPT Realtime costs $4.00 per 1M input tokens and $16.00 per 1M output tokens on OpenAI, with cache reads at $0.40 per 1M tokens. It has a 32,000-token context window and returns up to 4,096 tokens per response.
gpt-realtime.- Input
- $4.00 / 1M tokens
- Output
- $16.00 / 1M tokens
- Cache read
- $0.40 / 1M tokens
- Context window
- 32,000 tokens
- Max output
- 4,096 tokens
- Providers
- 2
- Added to LiteLLM
- September 2025
GPT Realtime pricing by provider
4 listings across 2 providers. Token prices are in US dollars per 1M tokens. The highlighted row is the price quoted above.
| Provider | Model name in LiteLLM | Input | Output | Cache read | Context |
|---|---|---|---|---|---|
| OpenAI | gpt-realtime | $4.00 | $16.00 | $0.40 | 32K |
| OpenAIsnapshot 2025-08-28 | gpt-realtime-2025-08-28 | $4.00 | $16.00 | $0.40 | 32K |
| Azure OpenAI | azure/gpt-realtime | $4.00 | $16.00 | $0.40 | 32K |
| Azure OpenAIsnapshot 2025-08-28 | azure/gpt-realtime-2025-08-28 | $4.00 | $16.00 | $0.40 | 32K |
Other GPT Realtime prices on OpenAI
| Audio input | $32.00 per 1M tokens |
| Audio output | $64.00 per 1M tokens |
What GPT Realtime costs in practice
1,000 requests with 2,000 input and 500 output tokens each
| OpenAI | $16.00 |
| Azure OpenAI | $16.00 |
GPT Realtime features
Input: text, image, audio. Output: text, audio.
- Function calling
- Parallel tool calls
- Audio input
- Audio output
Use GPT Realtime with LiteLLM
Call GPT Realtime through the LiteLLM Python SDK, or put it behind the LiteLLM proxy and send OpenAI-format requests to /v1/realtime. LiteLLM tracks the cost of every request at these prices.
from litellm import completion
response = completion(
model="gpt-realtime",
messages=[{"role": "user", "content": "Hello!"}],
)model_list:
- model_name: gpt-realtime
litellm_params:
model: gpt-realtime
api_key: os.environ/OPENAI_API_KEYcurl http://0.0.0.0:4000/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer sk-1234" \
-d '{
"model": "gpt-realtime",
"messages": [{"role": "user", "content": "Hello!"}]
}'GPT Realtime FAQ
How much does GPT Realtime cost?
GPT Realtime costs $4.00 per 1M input tokens and $16.00 per 1M output tokens on OpenAI, with cache reads at $0.40 per 1M tokens.
What is the context window of GPT Realtime?
GPT Realtime has a 32,000-token context window and returns up to 4,096 tokens per response.
Which providers offer GPT Realtime?
GPT Realtime is available from 2 providers through LiteLLM: OpenAI and Azure OpenAI.
How do I call GPT Realtime with an OpenAI-compatible API?
Use the LiteLLM Python SDK with model="gpt-realtime", or add the model to the LiteLLM proxy's config.yaml and send requests to /v1/realtime from any OpenAI client. LiteLLM tracks the cost of every request at the prices on this page.
What features does GPT Realtime support?
GPT Realtime supports function calling, parallel tool calls, audio input and audio output.
When did LiteLLM add GPT Realtime?
GPT Realtime was added to LiteLLM's model price file in September 2025.
Is GPT Realtime being deprecated?
OpenAI lists a deprecation date of January 20, 2027 for gpt-realtime.
Sources
Prices come from LiteLLM's open-source model_prices_and_context_window.json, which LiteLLM uses to calculate the cost of every request. Provider pricing pages:
- developers.openai.com/api/docs/pricing
- learn.microsoft.com/en-us/azure/foundry/foundry-models/concepts/models-sold-directly-by-azure
LiteLLM docs: OpenAI, Azure OpenAI.
Provider pages: OpenAI, Azure OpenAI.