Azure Realtime is ahead of Deepslate Opal on reliability, success rate, agent responses, data accuracy, stalled calls, response time and cost. The two tie on interruption.
Last updated Methodology by Dileep Chagam
Azure Realtime is the better pick over Deepslate Opal on every measure that separates them: reliability across repeated calls (13.4 points higher), calls that pass every check (14.2 points higher), getting the agent's replies right (2.8 points higher), saving the caller's data correctly (13.5 points higher), fewer stalled calls (1.2 points lower), fast replies (0.62s sooner) and lower cost ($0.032 cheaper). They are level on handling interruptions.
Scenarios that passed on all three runs.
Azure Realtime, 13.4 points higher
Calls that passed every check.
Azure Realtime, 14.2 points higher
Calls where the agent's replies reached the expected outcome.
Azure Realtime, 2.8 points higher
Calls where everything the agent saved matched exactly.
Azure Realtime, 13.5 points higher
Calls where the agent went silent for 10 seconds or more.
Azure Realtime, 1.2 points lower
How well it handled being talked over, 0 to 5.
Level
Median time to start replying after the caller finishes.
Azure Realtime, 0.62s sooner
One minute of call at the vendor's published rates.
Azure Realtime, $0.032 cheaper
The small figure is the place among ranked models. The cascade is not ranked.
Azure Realtime 1.66s, Deepslate Opal 2.27s at the median.
Azure Realtime is more reliable, by 13.4 points. Azure Realtime passed 58 of 82 scenarios on all three runs, and Deepslate Opal passed 47. A scenario only counts when every run of it passed.
For a phone agent, Azure Realtime, since it gets more scenarios right on every run. Azure Realtime is ahead of Deepslate Opal on reliability, success rate, agent responses, data accuracy, stalled calls, response time and cost. The two tie on interruption.
Azure Realtime, by 13.5 points. Azure Realtime saved everything exactly right on 86.1% of the calls that save data, and Deepslate Opal on 72.6%. On Appointments, 94.2% against 85.4%. On Medicare, 65.2% against 39.4%. A call counts only when nothing is missing, wrong, extra or saved twice.
Yes. Azure Realtime starts replying 0.62s sooner at the median. Azure Realtime takes 1.66s at the median and 2.22s at p90; Deepslate Opal takes 2.27s and 2.83s. Both are measured on the call audio, from the caller finishing to the agent starting to reply.
They score the same. Azure Realtime scores 4.96 and Deepslate Opal 4.96 for how well the agent handled being talked over.
Azure Realtime, by 1.2 points. The agent went silent for 10 seconds or more on 4.1% of Azure Realtime calls and 5.3% of Deepslate Opal calls.
Azure Realtime is $0.032 cheaper a minute. Azure Realtime costs $0.080 a minute of call and Deepslate Opal $0.112, 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, 2,952 calls across the benchmark. A call passes only when the agent said the right things and saved the right data.
