GPT-Realtime-2.1: API pricing, providers and specs
GPT-Realtime-2.1 is a audio model by OpenAI. 6 providers serve it, from $19.20 to $24 /1M; the lowest price is at CometAPI.
How GPT-Realtime-2.1 compares
Against the 43 audio models on the site priced the same way.
- Pricey for its class: at $19.20 /1M, its lowest price is above 86% of comparable models (median $6.15).
- Its 128K-token context window is larger than 39% of them (median 128K), about 190 pages of text.
- 6 providers serve it; 84% of comparable models have fewer.
GPT-Realtime-2.1 prices by provider
Every provider's list price, read from its own price page or API. A price marked “from” changes with context length, resolution or mode.
| Provider | Input | Output | Prompt caching | Context |
|---|---|---|---|---|
 CometAPI | $3.20/1M | $19.20/1M | $0.32/1M | — |
| Vercel AI Gateway | $4/1M | $24/1M | $0.40/1M | 128K |
 APIYI | $4/1M | $24/1M | — | — |
 AIHubMix | $4/1M | $24/1M | $0.40/1M | 128K |
| DevPass (LLM Gateway) | $4/1M | From$24/1M | $0.40/1M | — |
 FastRouter | $4/1M | $24/1M | $0.40/1M | 128K |
- CometAPI$19.20
- Vercel AI Gateway$24
- APIYI$24
- AIHubMix$24
- FastRouter$24
- DevPass (LLM Gateway)From$24
Specs
- Type
- Audio
- Maker
- OpenAI
- Inputs
- Text, Audio, Image
- Outputs
- Text, Audio
- Context
- 128K≈ 190 pages of text
- Max output
- 32K
- Capabilities
- Prompt caching, Reasoning, Tool calling, Vision
- Released
- Jul 9, 2026
- License
- proprietary
- Open weights
- No, API only
Alternatives to GPT-Realtime-2.1
Same kind of model from other makers, at a similar price.
| Model | Lowest price | Context | Arena | Providers |
|---|---|---|---|---|
| GPT-Realtime-2.1this model | $19.20/1M | 128K | — | 6 |
 Gemini 3.1 Flash TTS Preview | $14/1M | 8K | — | 10 |
 Qwen Audio 3.0 Realtime Plus | $6.40/1M | — | — | 1 |
| Gemini 2.5 Flash Lite Preview TTS | $10/1M | — | — | 1 |
| Gemini 2.5 Flash TTS | $10/1M | 8K | — | 4 |
| Gemini 2.5 Pro TTS | $14/1M | 8K | — | 5 |
| Gemini 3.5 Live Translate Preview | $21/1M | 131K | — | 1 |
| Gemini 3.5 Transcribe | $12/1M | — | — | 3 |
| Gemini 3.5 Transcribe Live | $21/1M | — | — | 2 |
More from OpenAI
GPT-Realtime-2.1 FAQ
The lowest price for GPT-Realtime-2.1 is $19.20 /1M, at CometAPI.
CometAPI, at $19.20 /1M. 6 providers serve GPT-Realtime-2.1; the table above lists every one with its price.
GPT-Realtime-2.1 reads up to 128K tokens per request and writes up to 32K.
No. GPT-Realtime-2.1 is only available through an API; OpenAI does not publish its weights.
6 API providers serve GPT-Realtime-2.1. Compare their prices above, or open a provider to see everything it offers.
Can't find what you're looking for? Contact us