OpenAI · Proprietary · Read 2026-10-27
OpenAI GPT-Realtime-2: latency, voice quality, price and languages
- Rank
- not ranked
- Speed tier
- n/a
- P95
- n/a
- Price / 1M chars
- $191.60
OpenAI GPT-Realtime-2 is a proprietary text-to-speech model from OpenAI. Coval does not measure its latency, so it has no speed tier and no rank. No source tested it in a language other than English.
This summary is generated from the measures below; it is not a vendor description. Measures a source does not hold are shown as n/a and are never estimated.
Speed (Coval, 30 days, US East, English)
| Speed tier | n/a |
| Median time to first audio | n/a |
| P90 | n/a |
| P95 | n/a |
| Mean | n/a |
| Standard deviation | n/a |
| Calls in the window | n/a |
| Cekura median, streamed text | n/a |
| Cekura median, full text | n/a |
Voice quality (never added together)
| Listener Elo (Artificial Analysis) | 1,073 ±12 |
| Customer-service Elo (Artificial Analysis) | 1,120 ±27 |
| Naturalness (Speko) | n/a |
| Naturalness estimate (Cekura, 1 to 5) | n/a |
Accuracy
| Reading accuracy (Cekura) | n/a |
| Numbers (Cekura) | n/a |
| Codes (Cekura) | n/a |
| Context (Cekura) | n/a |
| Medical terms (Cekura) | n/a |
| Hard passages judged correct (Artificial Analysis) | n/a |
| Numbers and dates (Speko score) | n/a |
| Word error rate (Coval) | n/a |
Price per 1 million characters
| Range across sources | $191.60 |
| Artificial Analysis | $191.60 |
| Speko | n/a |
| Coval | n/a |
| Cekura | n/a |
Coverage
| Access | Proprietary |
| Streaming | Vendor documentation |
| Coval tags | n/a |
| Rated for | Knowledge sharing, Assistants, Entertainment, Customer service |
| Languages tested | none beyond English |
Not collected
- Speed (Coval median time to first audio): not measured by Coval (reading of 2026-10-27, 30-day window)
- Speed (Coval P95 time to first audio): not measured by Coval (reading of 2026-10-27, 30-day window)
- Naturalness (Speko naturalness): not measured by Speko (reading of 2026-10-27)
- Numbers and dates (Speko pronunciation score): not measured by Speko (reading of 2026-10-27)
Sources
Keep exploring
Frequently asked questions
- How fast is OpenAI GPT-Realtime-2?
- Coval does not measure OpenAI GPT-Realtime-2, so no latency is published here.
- How much does OpenAI GPT-Realtime-2 cost?
- The benchmarks print a list price of $191.60 per million characters for OpenAI GPT-Realtime-2.
- Which languages was OpenAI GPT-Realtime-2 tested in?
- No source tested OpenAI GPT-Realtime-2 in a language other than English.