CekuraBench
Book a demo

GPT Realtime 2.1 vs GPT Realtime 2.1 Mini

All comparisonsLeaderboard
Compare
GPT Realtime 2.1OpenAI
Wins 9 of 11
Reliability (Pass³)
32.9%
Latency P50
2.16s
Cost per Minute (USD)
$0.085
GPT Realtime 2.1 MiniOpenAI
Wins 2 of 11
Reliability (Pass³)
14.3%
Latency P50
2.40s
Cost per Minute (USD)
$0.020

GPT Realtime 2.1 is ahead on ELO rating, hearing score, turn taking score, speaking score, task handling score, reliability, calls handled well, calls fully resolved and median latency, and GPT Realtime 2.1 Mini on calls where the agent went quiet and cost.

Last updated Methodology by Dileep Chagam

Who is better at what

GPT Realtime 2.1GPT Realtime 2.1 Mini
  • ELO Rating1054921
  • Hearing7254
  • Turn Taking7574
  • Speaking7971
  • Task Handling7865
  • Reliability (Pass³)32.9%14.3%
  • Pass Rate (score ≥ 80)56.7%39.5%
  • Task Completion Rate34.3%13.8%
  • Dead-Air Rate (≥ 10 s)11.9%9.0%
  • Latency P502.16s2.40s
  • Cost per Minute (USD)$0.085$0.020

Results by call

Right every time, on booking and Healthcare calls

Banking callsHealthcare calls
  1. 26%
    40%
    GPT Realtime 2.1
  2. 9%
    20%
    GPT Realtime 2.1 Mini

Related Comparisons

Each model on its own: GPT Realtime 2.1 and GPT Realtime 2.1 Mini

GPT Realtime 2.1 vs GPT Realtime 2.1 Mini FAQ

  • GPT Realtime 2.1 is more reliable, by 18.6 points. GPT Realtime 2.1 handled 23 of 70 real callers well on all three runs, and GPT Realtime 2.1 Mini 10. A caller only counts when every run of the call scored 80 or more.

  • For a voice agent, GPT Realtime 2.1, since it handles more real callers well on every run. GPT Realtime 2.1 is ahead on ELO rating, hearing score, turn taking score, speaking score, task handling score, reliability, calls handled well, calls fully resolved and median latency, and GPT Realtime 2.1 Mini on calls where the agent went quiet and cost.

  • GPT Realtime 2.1, by 20.5 points higher. GPT Realtime 2.1 fully resolved 34.3% of calls and GPT Realtime 2.1 Mini 13.8%. A call counts only when every step the caller needed was done.

  • Yes. GPT Realtime 2.1 starts replying 0.24s sooner at the median. GPT Realtime 2.1 takes 2.16s at the median and 2.70s at p90; GPT Realtime 2.1 Mini takes 2.40s and 2.88s. Both are measured on the call audio, from the caller finishing to the agent starting to reply.

  • GPT Realtime 2.1 Mini, by 2.9 points. The agent went quiet for 10 seconds or more on 11.9% of GPT Realtime 2.1 calls and 9.0% of GPT Realtime 2.1 Mini calls.

  • GPT Realtime 2.1 Mini is $0.065 cheaper a minute. GPT Realtime 2.1 costs $0.085 a minute of call and GPT Realtime 2.1 Mini $0.020, at each vendor's published rates.

  • 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.

Test your voice agent the same way.

Cekura runs these simulated calls on your agent and scores every one.

Book a demoStart free