Telnyx Launches Edge Compute to Complete the Real-Time AI Stack
The Agent Runtime joins Telnyx's inference, speech, and network, so an agent's code, memory, models, voice, and calls now run on one infrastructure, with zero hops between them.
Austin, TX, Aug. 03, 2026 (GLOBE NEWSWIRE) -- Telnyx today launched Telnyx Edge Compute, the Agent Runtime that lets an entire real-time voice AI agent run on infrastructure Telnyx owns.
The launch centers on the runtime, where an agent's own code lives. Functions run containerized code, deployed with one command. StatefulActor holds durable per-entity state, one instance per entity, one call at a time. KV, CloudFS, and Cloud Storage handle state, files, and objects. Functions reach state, storage, and inference through bindings, removing the need for credentials, network calls, or egress fees between components.
For years, Telnyx has run the other pieces a real-time agent needs. It hosts open-source LLMs on owned GPUs and runs speech-to-text and text-to-speech models on its platform, offering over 1400 voices. These capabilities, coupled with a private global telephony network, have powered real-time voice AI agents worldwide.
The one piece that still ran on someone else's setup was the runtime the agent's logic executes on. Edge Compute closes that gap. The agent's code now runs on the same infrastructure that hosts the models, the speech, and the calls, colocated on one network. That means zero hops across the whole pipeline, and everything a real-time voice agent needs runs entirely on Telnyx.
"The hard part of real-time AI is not the model. It is everything around it: the latency, the state, the network, the trust. Edge Compute is the layer that puts all of that on one infrastructure, and that is what makes agents production-ready instead of demo-ready," said David Casem, CEO, Telnyx.
Running on one infrastructure changes what agents can do. Responses are faster because compute, state, and inference sit together at sub-200ms round-trip. Operations are simpler because one API, one SLA, one bill, and end-to-end observability replace a stack of vendors and dashboards. Data stays in-region by architectural default, so agents are running in accordance with local and international regulations. This is the infrastructure that makes agents production-ready.
Edge Compute is available today via the Telnyx API and CLI. Users can explore live demos on the Edge Compute microsite, which itself runs on an Edge Compute Function.
About Telnyx
Telnyx is the infrastructure for real-time agents. Edge compute, voice AI, and global communications run on one system, on a carrier network Telnyx owns, with local numbers in 140+ countries and 18 global points of presence with colocated GPU infrastructure. That single operational domain delivers sub-200ms round-trip time, A-level STIR/SHAKEN authentication, and enterprise compliance including HIPAA, PCI, SOC 2, and GDPR, so AI agents are answered and trusted in production. Voice AI is where this converges first. The category is bigger: real-time AI infrastructure. Headquartered in Austin with offices worldwide. Learn more at telnyx.com.
Media Contact press@telnyx.com
Legal Disclaimer:
EIN Presswire provides this news content "as is" without warranty of any kind. We do not accept any responsibility or liability for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this article. If you have any complaints or copyright issues related to this article, kindly contact the author above.
