GPT-Realtime mini Dec 2025
GPT-Realtime mini places in the top three on 2 benchmarks.
7
Benchmarks with a score
0
Ranked first, out of five or more models
2
In the top three
OpenAI Dec 2025
GPT-Realtime mini
Dec 2025 snapshot Closed weights Small
No Epoch Capabilities Index score yet
- Audio MC 16.6% #17 /21
- vCX Benchmark 75% #38 /43
- Speko Speech-to-Speech Benchmark (10-turn call) 614% #5 /6
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Speko Speech-to-Speech Benchmark (six concierge scenarios) | 0.87 | #3 of 11 | Official leaderboard | |
| Speko Speech-to-Speech Benchmark (10-turn call) | 614% | #5 of 6 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| VISTA (Visually Impaired Scene and Task Annotations) | 3 | #2 of 11 | Official leaderboard | |
| VideoFDB | 2.73 | #4 of 5 | Official leaderboard | |
| Audio MC | 16.6% | #17 of 21 | Official leaderboard | |
| vCX Benchmark | 75% | #38 of 43 | Official leaderboard | |
| EVA-Bench | 0.16 | #11 of 11 | Official leaderboard |
Who measured it?
Independent runs cover all 7 of its benchmarks.
- An independent run only 7
Sources: Official leaderboards, Epoch AI (CC BY 4.0) and labs’ model cards