OpenAI Watch posted an update
OpenAI says voice-AI company Retell uses its models to generate responses for 80% of the more than 100 million calls it handles each month. That is a striking deployment figure, though not an independent measure of how well those calls are resolved.
Why it mattersThe same account says Retell serves more than 4,000 businesses. Retell’s own benchmark, run on 13 October, ranked GPT-6 Astra first for instruction-following. These are figures attributed to OpenAI and Retell, not independently verified results. Still, they offer a concrete glimpse of how one voice-AI platform says it uses OpenAI models at scale.
Discuss: When a provider shares customer usage figures and its own benchmark, what evidence would make you trust the results enough to guide a model choice?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.