Ringg's AI agents resolve 65% of customer calls using OpenAI models

Ringg, a voice and chat agent platform, has reported that its AI agents, powered by OpenAI's GPT-5.6 models, resolve up to 65% of customer inquiries without human intervention. The company claims that using GPT-5.6 reduces model costs by approximately 90% compared to GPT-4.1, while maintaining required quality and latency. Ringg's agents handle over 7 million connected calls monthly, with an average customer satisfaction score of 4.8. Ringg's platform uses a combination of GPT-5.6 models, including Luna for real-time interactions, Terra for post-call analysis, and Sol for evaluation workflows. The system dynamically routes tasks to the most suitable model based on the request's needs. Ringg also employs a context layer to maintain conversation continuity across voice, chat, WhatsApp, and web channels. The company's tests showed GPT-5.6 Terra outperformed alternatives like Gemini 2.5 Flash, achieving up to 97% accuracy in regional languages. Ringg's clients, such as Policybazaar and Practo, have seen significant improvements in response times and cost reductions. Policybazaar's average response time dropped from 8–12 minutes to under 60 seconds, while Practo's operating costs fell by 70% after deploying Ringg's agents. The company is also developing browser agents for onboarding, KYC, and claims processing, aiming to measure customer operations by automation depth rather than call volume.