OpenAI
GPT Realtime 2.1 Limits
Token limits
| Spec | Value |
|---|---|
| Context window | 128,000 tokens |
| Max output | 32,000 tokens |
| Input types | Text, Audio, Image |
| Output types | Text, Audio |
Rate limits
Per organization, by API usage tier.
| Tier | RPM | TPM |
|---|---|---|
| Build | 400 | 200,000 |
| Launch | 10,000 | 4,000,000 |
| Grow | 20,000 | 15,000,000 |
Source: openai.com, read Nov 9, 2026.
Get an email when we log a change to GPT Realtime 2.1
Questions
What is the GPT Realtime 2.1 context window
128,000 tokens, per developers.openai.com, read Nov 9, 2026.
What is the GPT Realtime 2.1 max output
32,000 tokens, per developers.openai.com.
What are the GPT Realtime 2.1 rate limits
Per organization, by API usage tier. RPM 400, TPM 200,000 at the Build tier, per developers.openai.com.
Source: openai.com, read Nov 9, 2026.