GPT Realtime Whisper
Frontier labTranscription
API id
gpt-realtime-whisperReleased 6 May 2026 (May 2026)Azure OpenAI (except 2026-05-06, 2026-05-07) retires it on 6 May 2027 (in 169 days). 2 hosts keep serving it.
Input / 1M
-
not listed
Output / 1M
-
not listed
Context window
32K
tokens
Max output
4K
tokens per reply
Hosts
2
4 listings
Released
May 2026
6 May 2026
Where to run it
2 hosts serve GPT Realtime Whisper. The comparison is on blended price, 3 parts input to 1 part output.
| Host | In / out per 1M | Model id | |
|---|---|---|---|
OpenAILab API gpt-realtime-whisper | - | gpt-realtime-whisper | |
Azure OpenAICloud gpt-realtime-whisper | - Retires 6 May 2027 | gpt-realtime-whisper | Retires 6 May 2027 |
Show 2 more listings: regions, snapshots and variantsHide regional listings
| Azure OpenAI | |||||||
| 2026-05-06 gpt-realtime-whisper-2026-05-06 | - | gpt-realtime-whisper-2026-05-06 | |||||
| 2026-05-07 gpt-realtime-whisper-2026-05-07 | - | gpt-realtime-whisper-2026-05-07 | |||||
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
- Audio
- Produces
- Text
- Features
- Audio input
Similar models
This model has no listed price, so these are current models with similar capabilities.
Smaller OpenAI models
No similar OpenAI model keeps its key capabilities.
Other frontier labs
Open-weight picks
No similar 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-whisper", input="Hello")
print(response.output_text)Shell
curl https://api.openai.com/v1/responses \
-H "Authorization: Bearer $OPENAI_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "gpt-realtime-whisper", "input": "Hello"}'Python
from litellm import completion
response = completion(
model="openai/gpt-realtime-whisper",
messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)Quick answers
Which APIs offer GPT Realtime Whisper?
GPT Realtime Whisper runs on 2 APIs: OpenAI and Azure OpenAI.
What is GPT Realtime Whisper's context window?
32K tokens, with up to 4K tokens in one reply.
Is GPT Realtime Whisper being retired?
Azure OpenAI (except 2026-05-06, 2026-05-07) retires it on 6 May 2027. 2 hosts keep serving it.
When was GPT Realtime Whisper released?
OpenAI released it on 6 May 2026.
What is GPT Realtime Whisper's API model id?
Use gpt-realtime-whisper with OpenAI.
Sources
Raw catalog entry for gpt-realtime-whisper on OpenAI
{
"input_cost_per_second": 0.000283333333333,
"litellm_provider": "openai",
"mode": "audio_transcription",
"source": "https://developers.openai.com/api/docs/pricing",
"supported_endpoints": [
"/v1/realtime",
"/v1/realtime/transcription_sessions"
],
"supported_modalities": [
"audio"
],
"supported_output_modalities": [
"text"
],
"supports_audio_input": true
}