PulseAugur
EN
LIVE 11:05:06

OpenAI launches GPT-Realtime-2.1-mini for low-latency voice agents

OpenAI has launched two new Realtime models, GPT-Realtime-2.1 and GPT-Realtime-2.1-mini, designed for low-latency voice and multimodal applications via their API. The GPT-Realtime-2.1-mini is a smaller, more cost-effective reasoning model for real-time voice interactions, maintaining previous pricing while adding enhanced reasoning and tool-use capabilities. OpenAI also announced a significant reduction in p95 latency by at least 25% across its Realtime voice models, attributed to improved caching mechanisms. AI

IMPACT Enhances real-time voice agent capabilities with improved latency and reasoning, potentially accelerating adoption in customer support and interactive applications.

RANK_REASON Frontier-lab model release with system card. [lever_c_demoted from frontier_release: ic=2 ai=1.0]

Read on MarkTechPost →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

OpenAI launches GPT-Realtime-2.1-mini for low-latency voice agents

COVERAGE [2]

  1. MarkTechPost TIER_1 English(EN) · Michal Sutter ·

    OpenAI Releases GPT-Realtime-2.1 and GPT-Realtime-2.1-mini for Low-Latency Voice Agents in the API

    <p>OpenAI added two Realtime models to its API. GPT-Realtime-2.1-mini is a mini reasoning model for voice, priced like the earlier gpt-realtime-mini. OpenAI also cut p95 latency by at least 25% through improved caching. Here is what changed, how pricing compares, and how to conne…

  2. Mastodon — mastodon.social TIER_1 English(EN) · [email protected] ·

    OpenAI has released GPT-Realtime-2.1-mini, a compact reasoning model for voice interactions. It keeps the earlier mini pricing while adding reasoning, and OpenA

    OpenAI has released GPT-Realtime-2.1-mini, a compact reasoning model for voice interactions. It keeps the earlier mini pricing while adding reasoning, and OpenAI cut p95 latency by at least 25 percent through improved caching. https://www. marktechpost.com/2026/07/06/op enai-gpt-…