Lowest price
Either one
Both cost $32.00 / 1M audio tokens.
GPT-Realtime vs GPT-Realtime-2.1: price, context and capabilities
Choose GPT-Realtime-2.1 for very long inputs, the latest release and the most recent knowledge; 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).
- GPT-Realtime-2.1 reads more per request: 128k tokens against 32k.
- GPT-Realtime-2.1 is the newer model (Jul 6, 2026, against Aug 28, 2025).
- 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.
Very long inputs
Choose GPT-Realtime-2.1
GPT-Realtime-2.1 reads up to 128k tokens per request, against 32k for GPT-Realtime.
Latest release
Choose GPT-Realtime-2.1
GPT-Realtime-2.1 is the newer release (Jul 6, 2026, against Aug 28, 2025 for GPT-Realtime).
Most recent knowledge
Choose GPT-Realtime-2.1
GPT-Realtime-2.1 was trained on data up to September 2024, against October 2023 for GPT-Realtime.
Specifications side by side
| GPT-Realtime | GPT-Realtime-2.1 | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Category | Voice and realtime | Voice and realtime |
| Status | Deprecated | Generally available |
| Headline price | $32.00 / 1M audio tokens | $32.00 / 1M audio tokens |
| Context window | 32k | 128k |
| Max output | 4k | 32k |
| Released | Aug 28, 2025 | Jul 6, 2026 |
| Knowledge cutoff | October 2023 | September 2024 |
| Licence | Proprietary | Proprietary |
| Capabilities | Audio input, Tool calling | Audio input, Reasoning, Tool calling |
| API model id | gpt-realtime | gpt-realtime-2.1 |
Frequently asked questions
Should I choose GPT-Realtime or GPT-Realtime-2.1?
Choose GPT-Realtime-2.1 for very long inputs, the latest release and the most recent knowledge; on the other points the two are level or the data is missing.
Which is cheaper, GPT-Realtime 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-Realtime or GPT-Realtime-2.1?
GPT-Realtime-2.1 reads more per request: 128k tokens against 32k.
What is the difference between GPT-Realtime and GPT-Realtime-2.1?
GPT-Realtime is made by OpenAI and GPT-Realtime-2.1 by OpenAI. Both cost the same at the headline price ($32.00 / 1M audio tokens). GPT-Realtime-2.1 reads more per request: 128k tokens against 32k. GPT-Realtime-2.1 is the newer model (Jul 6, 2026, against Aug 28, 2025). Only GPT-Realtime-2.1 offers reasoning.
Which is newer, GPT-Realtime or GPT-Realtime-2.1?
GPT-Realtime-2.1 was released on Jul 6, 2026 and GPT-Realtime on Aug 28, 2025.