CekuraBench

Deepslate Opal

Every modelCompare

57.3%

Reliability

8th of 12

2.27s

Median reply

11th of 12

$0.112

Cost / min

8th of 9

Reliability vs. response time

Higher and further left is better

Best quadrant
Reliability, pass³ (%)
Median response time (s)

For the wider field, see our pick of the best voice AI APIs for real-time audio.

Deepslate Opal against every other model

Scenarios that passed on all 3 runs, pass³. Higher is better.

#ModelReliability
  1. 01Grok Voice 3.081.7%
  2. 02GPT Realtime 2.179.3%
  3. 03GPT-Live 1 + GPT-6 Sol75.6%
  4. 04Azure Realtime70.7%
  5. 05Phonic v168.3%
  6. 06Grok Voice Think Fast 2.067.1%
  7. 07Gemini 3.1 Flash Live Preview62.2%
  8. 08Deepslate Opal57.3%
  9. 09Gemini 3.8 Live Extended Thinking52.4%
  10. 10Amazon Nova 2 Sonic50.0%
  11. 11GPT Realtime 2.1 Mini50.0%
  12. 12Gemini 3.8 Live41.5%
  • refCascade82.9%

Where Deepslate Opal struggled

Deepslate Opal falls furthest behind on silence, cough and packet loss: 58% passed, against a median of 88%.

Below the medianPassedMedian
  • Silence, cough & packet loss
    58%median 88%
  • Accent & slow speech
    67%median 83%
  • Background noise
    67%median 81%
  • Medicare compliance & routing
    30%median 40%
  • Tool or system errors
    83%median 92%
  • Caller corrects or changes mind
    79%median 81%
  • Plain booking & lookup
    93%median 94%
Mistakes in saved data
  • Step skipped58 · 40%
  • Wrong value45 · 31%
  • Extra call18 · 12%

Fields it got wrong most: phone, route_id, callback_phone

Response time at P50, P90 and P95

Deepslate Opal starts replying in 2.27s on a typical turn and 3.15s on the slowest turn in twenty, 11th fastest of 12 at the median.

ModelP50P90P95
  • Cascade
    2.17s2.51s2.62s
Seconds from the caller finishing to the agent replying. The bar is the typical turn, P50.

Deepslate Opal head to head

Every comparisonAll 12 models

How much audio Deepslate Opal was tested on

246 live calls: 82 scenarios, each run three times. The median call ran 1:44.

Live calls
246
Scenarios
82
Booking calls
177
Medicare calls
69
Replies per call
8.0
Median call
1:44

82 scenarios, the same for every model. See every test case

Deepslate Opal FAQ

  • Deepslate Opal passed 47 of 82 scenarios on all three runs, 57.3%, 8th of 12 realtime models. Across all 246 calls, 69.9% passed every check.

  • Deepslate Opal starts replying 2.27s after the caller finishes at the median, 2.83s at P90 and 3.15s at P95. That is 11th fastest of 12 at the median, measured on the call audio.

  • Deepslate Opal falls furthest behind the field on silence, cough and packet loss: 58% of calls passed, against a median of 88%. When it saved data wrong, the most common mistake was step skipped, 40% of them.

  • Deepslate Opal saved everything exactly right on 72.6% of the calls that save data, 9th of 12. On Booking calls, 85.4%. On Medicare calls, 39.4%.

  • $0.112 a minute of call at Deepslate's published rates, 8th cheapest of 9 priced models. The vendor's regular rate of €0.10 a minute, at the ECB rate of 5 Oct 2026, applied to each call's length; its launch offer is €0.02. Excludes the ElevenLabs voice, billed on the customer's own ElevenLabs key.

  • Cekura ran Deepslate Opal as the whole agent in the same open-source Pipecat pipeline as every other model, with the same prompt, tools and connection. A simulated caller worked through 82 scenarios on live calls, three times each: 246 calls for Deepslate Opal. A call passes only when the agent said the right things and saved the right data.