GPT-Realtime mini: API pricing, providers and specs
GPT-Realtime mini is a audio model by OpenAI. 3 providers serve it, from $2.40 to $2.40 /1M; the lowest price is at Vercel AI Gateway.
How GPT-Realtime mini compares
Against the 43 audio models on the site priced the same way.
- Cheap for its class: at $2.40 /1M, its lowest price undercuts 75% of comparable models (median $6.15).
- Its 32K-token context window is larger than 30% of them (median 128K), about 50 pages of text.
- 3 providers serve it; 59% of comparable models have fewer.
GPT-Realtime mini 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 |
|---|---|---|---|---|
| Vercel AI Gateway | $0.60/1M | $2.40/1M | $0.06/1M | — |
| DevPass (LLM Gateway) | $0.60/1M | From$2.40/1M | $0.06/1M | — |
 FastRouter | $0.60/1M | $2.40/1M | $0.06/1M | 32K |
- Vercel AI Gateway$2.40
- FastRouter$2.40
- DevPass (LLM Gateway)From$2.40
Specs
- Type
- Audio
- Maker
- OpenAI
- Inputs
- Text, Audio
- Outputs
- Text, Audio
- Context
- 32K≈ 50 pages of text
- Max output
- 4K
- Capabilities
- Prompt caching, Tool calling
- Released
- Oct 10, 2025
- License
- —
- Open weights
- —
Alternatives to GPT-Realtime mini
Same kind of model from other makers, at a similar price.
| Model | Lowest price | Context | Arena | Providers |
|---|---|---|---|---|
| GPT-Realtime minithis model | $2.40/1M | 32K | — | 3 |
 Gemini 3.8 Flash TTS | $6.30/1M | 8K | — | 8 |
| Gemini TTS | $1/1M | — | — | 1 |
| Gemini 3.8 Live | $4.50/1M | 131K | — | 3 |
| Gemini 3.8 Flash-Lite TTS | $4.20/1M | 8K | — | 7 |
 Qwen Audio 3.1 ASR | $0.94/1M | — | — | 1 |
| Qwen Audio 3.1 TTS | $3.74/1M | — | — | 1 |
| Qwen Audio 3.0 Realtime Plus | $6.40/1M | — | — | 1 |
| Qwen Audio 3.1 ASR Message | $1.40/1M | — | — | 1 |
More from OpenAI
GPT-Realtime mini FAQ
The lowest price for GPT-Realtime mini is $2.40 /1M, at Vercel AI Gateway.
Vercel AI Gateway, at $2.40 /1M. 3 providers serve GPT-Realtime mini; the table above lists every one with its price.
GPT-Realtime mini reads up to 32K tokens per request and writes up to 4K.
We have not found GPT-Realtime mini's weights published by its maker.
3 API providers serve GPT-Realtime mini. Compare their prices above, or open a provider to see everything it offers.
Can't find what you're looking for? Contact us