Enterprise voice AI now supports local environment deployment

ElevenLabs is expanding its range of deployment options optimized for enterprise environments, beyond cloud and VPC.

• On-premise: Runs on GPU-based Confidential Computing infrastructure in a company's own servers or data centers. Suitable for government agencies or specific organizations where procuring cloud infrastructure is difficult.
• On-device: Runs directly on hardware, enabling offline inference even in limited computing environments. Can be used for in-vehicle voice systems by automakers or wearable devices.
• VPC (Virtual Private Cloud): Provided on AWS SageMaker and GCP Vertex environments for organizations where data security is important, and is currently available immediately.

These models are not simply packaged versions of the cloud model, but dedicated models designed for each environment. The On-premise and On-device models also support Fine-tuning for custom voices and specific languages/dialects.

On-premise and On-device are currently in early access, with general availability targeted for the first half of 2026.