Compare speech-to-speech models on live calls
Publishes benchmarks for nine realtime voice models tested as complete phone agents across 82 scenarios, with three runs per scenario.
Loading page…
Audio and video / product dossier
Cekura Bench
Benchmarks realtime voice models on live phone calls, with public transcripts and reliability metrics.
Product brief
What Cekura Bench does.
Cekura Bench publishes voice AI benchmarks you can verify. Our new speech-to-speech benchmark tests 9 realtime voice models, including GPT Realtime 2.1, Gemini Live, Grok and Phonic, as complete phone agents on live calls: 82 scenarios, three runs each. Models are ranked on reliability, data accuracy, stalled calls, response time and cost, and every call transcript is public. Cekura Bench also covers voice agent benchmarks and STT benchmarks, with TTS benchmarks coming soon.