GPT-Live Models by OpenAI
OpenAIProprietary
1 model2026
Last refreshed 2026-09-10. Next refresh: weekly.
Details
ResearcherOpenAI
LicenseProprietary
Commercial useCommercial use: conditional
Models1
Released2026
Capabilities
MultimodalAll models
JSON / Tool useAll models
Links
WebsiteAbout
GPT-Live is OpenAI's full-duplex voice model family for real-time conversations via the Live API. Models listen and speak simultaneously and can delegate reasoning and tool use to a backend agent.
Decision facts
- Best fit
- voicevision and multimodal workJSON / Tool use
- Capability starting point
- GPT-Live-1 with JSON / Tool use and multimodal inputs
- Lowest tracked input
- Not tracked
- Closest related family
- GPT Realtime 2
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
1 in view
GPT-Live-1Current
Use when the workload needs voice, JSON / Tool use, and multimodal inputs.
2026-07voiceJSON / Tool usemultimodal inputs
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| GPT-Live-1 | Use when the workload needs voice, JSON / Tool use, and multimodal inputs. | 2026-07 | voiceJSON / Tool usemultimodal inputs | Current |
Release Timeline
1 release group2026-07
1 current
GPT-Live-1
CurrentvoiceJSON / Tool usemultimodal inputs
Specifications(1 models)
| Model | Released | Multimodal | JSON / Tool use |
|---|---|---|---|
| GPT-Live-1 | 2026-07 | Yes | Yes |
