Flux → GPT-4.1 → ElevenLabs Flash is ahead on reliability, success rate, agent responses, data accuracy and stalled calls, and GPT Realtime 2.1 Mini on interruption and response time.
Scenarios that passed on all three runs.
Cascade, GPT-4.1, 32.9 points higher
Calls that passed every check.
Cascade, GPT-4.1, 31.7 points higher
Calls where the agent's replies reached the expected outcome.
Cascade, GPT-4.1, 5.7 points higher
Calls where everything the agent saved matched exactly.
Cascade, GPT-4.1, 29.6 points higher
Calls where the agent went silent for 10 seconds or more.
Cascade, GPT-4.1, 1.6 points lower
How well it handled being talked over, 0 to 5.
Realtime 2.1 Mini, 0.03 higher
Median time to start replying after the caller finishes.
Realtime 2.1 Mini, 0.09s 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, Cascade, GPT-4.1 2.17s at the median.
GPT Realtime 2.1 Mini passed 41 of 82 scenarios on all three runs, and Flux → GPT-4.1 → ElevenLabs Flash passed 68. Flux → GPT-4.1 → ElevenLabs Flash is ahead, by 32.9 points.
GPT Realtime 2.1 Mini saved everything exactly right on 64.5% of the calls that save data, and Flux → GPT-4.1 → ElevenLabs Flash on 94.1%. On Appointments, 88.8% against 98.8%. On Medicare, 0.0% against 81.8%. A call counts only when nothing is missing, wrong, extra or saved twice.
GPT Realtime 2.1 Mini starts replying 0.09s sooner at the median. GPT Realtime 2.1 Mini takes 2.08s at the median and 2.75s at p90; Flux → GPT-4.1 → ElevenLabs Flash takes 2.17s and 2.51s. 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 Flux → GPT-4.1 → ElevenLabs Flash, 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. The cascade runs the same agent for reference and is never ranked.
