GPT Realtime 2 Models by OpenAI
Last refreshed 2026-10-19. Next refresh: weekly.
Capabilities
Links
WebsiteAbout
GPT Realtime 2 is OpenAI's second generation real-time voice model family, released May 7, 2026. The family includes gpt-realtime-2 for voice reasoning, gpt-realtime-translate for live speech-to-speech translation, and gpt-realtime-whisper for streaming speech-to-text transcription. These models use distinct Realtime API endpoints and supersede the GPT-4o Realtime Preview generation for voice-agent workflows.
Decision facts
- Best fit
- realtime voicetranslationspeech recognition
- Capability starting point
- GPT Realtime 2 with 128K context and reasoning, JSON / Tool use, and multimodal inputs
- Lowest tracked input
- GPT Realtime 2.1 Mini · $0.600/1M · OpenAI API
- Closest related family
- GPT-5
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
Use when the workload needs realtime voice, 128K context, and reasoning.
Use when the workload needs realtime voice, 128K context, and reasoning.
Use when the workload needs realtime voice, 128K context, and reasoning.
Use when the workload needs translation, multimodal inputs, and audio.
Use when the workload needs speech recognition and audio.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| GPT Realtime 2.1 | Use when the workload needs realtime voice, 128K context, and reasoning. | 2026-07 | realtime voice128K contextreasoning | Current |
| GPT Realtime 2.1 Mini | Use when the workload needs realtime voice, 128K context, and reasoning. | 2026-07 | realtime voice128K contextreasoning | Current |
| GPT Realtime 2 | Use when the workload needs realtime voice, 128K context, and reasoning. | 2026-05 | realtime voice128K contextreasoning | Current |
| GPT Realtime Translate | Use when the workload needs translation, multimodal inputs, and audio. | 2026-05 | translationmultimodal inputsaudio | Current |
| GPT Realtime Whisper | Use when the workload needs speech recognition and audio. | 2026-05 | speech recognitionaudio | Current |
Release Timeline
2 release groupsSpecifications(5 models)
| Model | Released | Context | Vision | Multimodal | Reasoning | JSON / Tool use |
|---|---|---|---|---|---|---|
| GPT Realtime 2.1 | 2026-07 | 128K | Yes | Yes | Yes | No |
| GPT Realtime 2.1 Mini | 2026-07 | 128K | Yes | Yes | Yes | No |
| GPT Realtime 2 | 2026-05 | 128K | No | Yes | Yes | Yes |
| GPT Realtime Translate | 2026-05 | — | No | Yes | No | No |
| GPT Realtime Whisper | 2026-05 | — | No | No | No | No |
Available From(3 providers)
Pricing
| Model | Provider | Input / 1M | Output / 1M | Type |
|---|---|---|---|---|
| GPT Realtime 2.1 Mini | OpenAI API | $0.6 | $2.4 | Serverless |
| GPT Realtime 2.1 Mini | Microsoft Foundry | $0.6 | $2.4 | Serverless |
| GPT Realtime 2.1 | OpenAI API | $4 | $24 | Serverless |
| GPT Realtime 2.1 | Vercel AI Gateway | $4 | $24 | Serverless |
| GPT Realtime 2.1 | Microsoft Foundry | $4 | $24 | Serverless |
| GPT Realtime 2 | OpenAI API | $32 | $64 | Serverless |
