GPT-Realtime-1.5
2026-02-23 · Publisher : OpenAI · Domain : Artificial Intelligence
OpenAI releases GPT-Realtime-1.5 in the Realtime API, its flagship audio model for voice agents and customer support, with gpt-audio-1.5 for the Chat Completions API.
What changes
- Native speech-to-speech model, with no transcription, text and speech synthesis chain.
- Text, audio and image input; text and audio output.
- gpt-audio-1.5 brings audio input and output to the Chat Completions API.
- Successor to the gpt-realtime generation, followed in May by GPT-Realtime-2.
### Specifications - GPT-Realtime-1.5: 32,000-token context, maximum output 4,096 tokens, knowledge cutoff September 30, 2024. - gpt-audio-1.5: 128,000-token context, maximum output 16,384 tokens. - Availability: Realtime API (v1/realtime) and Chat Completions. - Identifiers: gpt-realtime-1.5, gpt-audio-1.5. ### Pricing | Model | Modality | Input | Cached input | Output | | --- | --- | --- | --- | --- | | gpt-realtime-1.5 | Text | $4.00 | $0.40 | $16.00 | | gpt-realtime-1.5 | Audio | $32.00 | $0.40 | $64.00 | | gpt-realtime-1.5 | Image | $5.00 | $0.50 | n/a | | gpt-audio-1.5 | Text | $2.50 | n/a | $10.00 | | gpt-audio-1.5 | Audio | $32.00 | n/a | $64.00 | Price per million tokens (USD), read from the model pages. ### Benchmarks No numerical score published by the vendor in the announcement. ### Key takeaways An update to the API's voice layer, with no dedicated post. It mainly serves as a point of comparison with GPT-Realtime-2 and 2.1.
Links
Tags : Audio, OpenAI, Realtime, Voix, Agent vocal