Lowest price
Either one
Both cost $32.00 / 1M audio tokens.
GPT-Audio-1.5 vs GPT-Realtime-2.1: price, context and capabilities
Choose GPT-Realtime-2.1 for the latest release; on the other points the two are level or the data is missing.
In short
- Both cost the same at the headline price ($32.00 / 1M audio tokens).
- Both accept 128k tokens per request.
- GPT-Realtime-2.1 is the newer model (Jul 6, 2026, against Feb 23, 2026).
- Only GPT-Realtime-2.1 offers reasoning.
Which should you choose?
Each answer below comes straight from the published prices, specifications and independent benchmark results.
Latest release
Choose GPT-Realtime-2.1
GPT-Realtime-2.1 is the newer release (Jul 6, 2026, against Feb 23, 2026 for GPT-Audio-1.5).
Specifications side by side
| GPT-Audio-1.5 | GPT-Realtime-2.1 | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Category | Voice and realtime | Voice and realtime |
| Status | Generally available | Generally available |
| Headline price | $32.00 / 1M audio tokens | $32.00 / 1M audio tokens |
| Context window | 128k | 128k |
| Max output | 16k | 32k |
| Released | Feb 23, 2026 | Jul 6, 2026 |
| Knowledge cutoff | September 2024 | September 2024 |
| Licence | Proprietary | Proprietary |
| Capabilities | Audio input, Tool calling | Audio input, Reasoning, Tool calling |
| API model id | gpt-audio-1.5 | gpt-realtime-2.1 |
Frequently asked questions
Should I choose GPT-Audio-1.5 or GPT-Realtime-2.1?
Choose GPT-Realtime-2.1 for the latest release; on the other points the two are level or the data is missing.
Which is cheaper, GPT-Audio-1.5 or GPT-Realtime-2.1?
Both cost the same at the headline price ($32.00 / 1M audio tokens).
Which has the larger context window, GPT-Audio-1.5 or GPT-Realtime-2.1?
Both accept 128k tokens per request.
What is the difference between GPT-Audio-1.5 and GPT-Realtime-2.1?
GPT-Audio-1.5 is made by OpenAI and GPT-Realtime-2.1 by OpenAI. Both cost the same at the headline price ($32.00 / 1M audio tokens). Both accept 128k tokens per request. GPT-Realtime-2.1 is the newer model (Jul 6, 2026, against Feb 23, 2026). Only GPT-Realtime-2.1 offers reasoning.
Which is newer, GPT-Audio-1.5 or GPT-Realtime-2.1?
GPT-Realtime-2.1 was released on Jul 6, 2026 and GPT-Audio-1.5 on Feb 23, 2026.