Cekura S2S - Success Rate: leaderboard
Cekura Bench realtime speech-to-speech leaderboard: each voice model runs as the agent on 82 scripted phone scenarios (59 patients booking, moving or cancelling clinic visits; 23 Medicare callers whose eligibility is checked and who are routed), three times each, inside the same open-source Pipecat agent with the same prompt and tools, over live WebRTC audio including noisy and glitchy calls. Labels carry the model setting Cekura states. Score: success rate (%) over all 246 calls, where a call succeeds when the agent said the right things and saved the right details. Higher is better.
Metric: Success rate (%) over all 246 calls. Source: benchmarks.cekura.ai. Saturation forecast: Rough model projection: around 2026. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Phonic 1 (Intelligence high) | 95.1 | |
| 2 | Cekura Cascade (flux-general-en → gpt-4.1 → eleven_flash_v2_5) | 93.1 | |
| 3 | GPT Realtime 2.1 (Reasoning high) | 89 | |
| 4 | Grok Voice 3.0 (Reasoning high) | 86.6 | |
| 5 | Azure Realtime | 84.1 | |
| 6 | GPT-Live 1 + GPT-6 Sol (Sol on low thinking) | 83.7 | |
| 7 | Grok Voice Think Fast 2.0 (Reasoning high) | 76 | |
| 8 | Gemini 3.1 Flash Live Preview (Thinking level high) | 76 | |
| 9 | Gemini 3.8 Live Extended Thinking (Thinking level high) | 76 | |
| 10 | Deepslate Opal | 69.9 |
Interactive version: theaggregate.ai/benchmark?slug=cekura-s2s-success-rate · How It Works · Data refreshed daily, snapshot 2026-11-05.