GPT Realtime 2.1 Mini is ahead on success rate, data accuracy and response time, and Amazon Nova 2 Sonic on agent responses and stalled calls. The two tie on reliability and interruption.
Scenarios that passed on all three runs.
Level
Calls that passed every check.
Realtime 2.1 Mini, 2.0 points higher
Calls where the agent's replies reached the expected outcome.
Nova 2 Sonic, 0.4 points higher
Calls where everything the agent saved matched exactly.
Realtime 2.1 Mini, 4.6 points higher
Calls where the agent went silent for 10 seconds or more.
Nova 2 Sonic, 1.2 points lower
How well it handled being talked over, 0 to 5.
Level
Median time to start replying after the caller finishes.
Realtime 2.1 Mini, 0.10s sooner
One minute of call at the vendor's published rates.
Not comparable
The small figure is the place among ranked models. The cascade is not ranked.
Realtime 2.1 Mini 2.08s, Nova 2 Sonic 2.18s at the median.
Amazon Nova 2 Sonic passed 41 of 82 scenarios on all three runs, and GPT Realtime 2.1 Mini passed 41. That is level.
Amazon Nova 2 Sonic saved everything exactly right on 59.9% of the calls that save data, and GPT Realtime 2.1 Mini on 64.5%. On Appointments, 82.5% against 88.8%. On Medicare, 1.5% against 0.0%. A call counts only when nothing is missing, wrong, extra or saved twice.
GPT Realtime 2.1 Mini starts replying 0.10s sooner at the median. Amazon Nova 2 Sonic takes 2.18s at the median and 2.76s at p90; GPT Realtime 2.1 Mini takes 2.08s and 2.75s. Both are measured on the call audio, from the caller finishing to the agent starting to reply.
There is no verified per-minute rate for Amazon Nova 2 Sonic, so cost cannot be compared yet.
Both ran as the whole agent in the same open-source Pipecat pipeline, with the same prompt, tools and phone line. A simulated caller worked through all 82 scenarios on live calls, three times each, 2,214 calls across the benchmark. A call passes only when the agent said the right things and saved the right data.
