GPT-Live 1 + GPT-6 Sol vs Rime Prism

GPT-Live 1 + GPT-6 Sol is more reliable, by 7.3 points. GPT-Live 1 + GPT-6 Sol passed 62 of 82 scenarios on all three runs, and Rime Prism passed 56. A scenario only counts when every run of it passed.

For a voice agent, GPT-Live 1 + GPT-6 Sol, since it gets more scenarios right on every run. GPT-Live 1 + GPT-6 Sol is ahead on reliability, success rate, agent responses and data accuracy, and Rime Prism on calls where the agent went quiet, interruption and response time.

GPT-Live 1 + GPT-6 Sol, by 4.6 points. GPT-Live 1 + GPT-6 Sol saved everything exactly right on 89.0% of the calls that save data, and Rime Prism on 84.5%. On Booking calls, 100.0% against 92.4%. On Medicare calls, 60.6% against 64.2%. A call counts only when nothing is missing, wrong, extra or saved twice.

No. Rime Prism starts replying 0.68s sooner at the median. GPT-Live 1 + GPT-6 Sol takes 2.02s at the median and 2.42s at p90; Rime Prism takes 1.34s and 1.57s. Both are measured on the call audio, from the caller finishing to the agent starting to reply.

Rime Prism, by 0.02 on a 0 to 5 scale. GPT-Live 1 + GPT-6 Sol scores 4.89 and Rime Prism 4.91 for how well the agent handled being talked over.

Rime Prism, by 12.2 points. The agent went quiet for 10 seconds or more on 12.2% of GPT-Live 1 + GPT-6 Sol calls and 0.0% of Rime Prism calls.

There is no verified per-minute rate for Rime Prism, so cost cannot be compared yet.

Cekura ran both as the whole agent in the same open-source Pipecat pipeline, with the same prompt, tools and connection. A simulated caller worked through all 82 scenarios on live calls, three times each, 3,690 calls across the benchmark. A call passes only when the agent said the right things and saved the right data.