GPT-Realtime-2
GPT-Realtime-2 places in the top three on Big Bench Audio.
7
Benchmarks with a score
0
Ranked first, out of five or more models
1
In the top three
OpenAI May 2026
GPT-Realtime-2
Closed weights Flagship
No Epoch Capabilities Index score yet
- Audio MC 48.5% #6 /18
- τ-Voice 42.4% #5 /7
- Realtime-Bench 15.3% #5 /8
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 | |
|---|---|---|---|---|
| Realtime-Bench | 15.3% | #5 of 8 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Audio MC | 48.5% | #6 of 18 | Official leaderboard | |
| Big Bench Audio | 96.6% | #3 of 7 | Lab model card | |
| vCX Benchmark | 83.9% | #26 of 42 | Official leaderboard | |
| EVA-Bench | 0.50 | #5 of 8 | Official leaderboard | |
| τ-Voice | 42.4% | #5 of 7 | Official leaderboard | |
| Audio Arena | 71.8 | #6 of 6 | Official leaderboard |
Its lab's number next to a leaderboard
Where the benchmark's own leaderboard, or Epoch AI's run, also has this model. Gaps mostly come from different harnesses, effort levels and task subsets. Every pair.
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.