OpenAI: GPT Realtime Translate API
GPT Realtime Translate is a streaming speech-to-speech translation model for live multilingual audio experiences. It uses a dedicated realtime translation endpoint and returns translated audio plus transcript deltas while source audio is still arriving. GPT Realtime Translate is priced by audio duration rather than text tokens.
- Context window: 16,000 tokens
- Max output: 2,000 tokens
- Input: audio
- Output: text
- Released: 2026-05-06
- Knowledge cutoff: 2024-09-29
Frequently Asked Questions
What is the context window of GPT Realtime Translate?
GPT Realtime Translate supports a context window of up to 16,000 tokens.
What is the knowledge cutoff of GPT Realtime Translate?
The knowledge cutoff of GPT Realtime Translate is 2024-09-29.