Higgs Realtime vs GPT Realtime 2.1

GPT Realtime 2.1 is more reliable, by 23.2 points. Higgs Realtime passed 46 of 82 scenarios on all three runs, and GPT Realtime 2.1 passed 65. A scenario only counts when every run of it passed.

For a voice agent, GPT Realtime 2.1, since it gets more scenarios right on every run. GPT Realtime 2.1 is ahead on reliability, success rate, agent responses and data accuracy, and Higgs Realtime on calls where the agent went quiet, interruption and cost. The two tie on response time.

GPT Realtime 2.1, by 25.7 points. Higgs Realtime saved everything exactly right on 66.2% of the calls that save data, and GPT Realtime 2.1 on 92.0%. On Booking calls, 91.8% against 95.9%. On Medicare calls, 0.0% against 81.8%. A call counts only when nothing is missing, wrong, extra or saved twice.

They start replying at the same median time. Higgs Realtime takes 1.94s at the median and 2.02s at p90; GPT Realtime 2.1 takes 1.94s and 2.46s. Both are measured on the call audio, from the caller finishing to the agent starting to reply.

Higgs Realtime, by 0.02 on a 0 to 5 scale. Higgs Realtime scores 5.00 and GPT Realtime 2.1 4.98 for how well the agent handled being talked over.

Higgs Realtime, by 2.0 points. The agent went quiet for 10 seconds or more on 1.2% of Higgs Realtime calls and 3.3% of GPT Realtime 2.1 calls.

Higgs Realtime is $0.0792 cheaper a minute. Higgs Realtime costs $0.0061 a minute of call and GPT Realtime 2.1 $0.0853, at each vendor's published rates.

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.