GPT Realtime 2.1 Mini vs Cascade

Cascade is more reliable, by 11.4 points. GPT Realtime 2.1 Mini handled 10 of 70 real callers well on all three runs, and Cascade 18. A caller only counts when every run of the call scored 80 or more.

On reliability, no: the cascade handles more callers well on every run. Cascade is ahead on ELO rating, speaking score, reliability, calls handled well, calls where the agent went quiet and median latency, and GPT Realtime 2.1 Mini on hearing score, turn taking score and calls fully resolved. The two tie on task handling score. Both ran as the same agent, with the same prompt and tools.

GPT Realtime 2.1 Mini, by 1.4 points higher. GPT Realtime 2.1 Mini fully resolved 13.8% of calls and Cascade 12.4%. A call counts only when every step the caller needed was done.

No. Cascade starts replying 0.34s sooner at the median. GPT Realtime 2.1 Mini takes 2.40s at the median and 2.88s at p90; Cascade takes 2.06s and 2.49s. Both are measured on the call audio, from the caller finishing to the agent starting to reply.

Cascade, by 1.9 points. The agent went quiet for 10 seconds or more on 9.0% of GPT Realtime 2.1 Mini calls and 7.1% of Cascade calls.

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

Cekura ran both as the whole agent with the same prompt, tools and phone connection. Both took the same 70 recorded real human callers on live calls, three times each. Each call is graded on 25 skills across Hearing, Turn Taking, Task Handling and Speaking, and passes at a score of 80 out of 100. Cascades run the same agent for reference and are never ranked.