Roadmap

This page highlights upcoming features and improvements for Privatemode. We strive to provide the most secure and best performing confidential AI inference platform.

Privatemode web

Currently available via API and desktop applications, Privatemode is expanding to the web. The upcoming web launch will deliver a seamless, ChatGPT-like user experience directly in the browser, while upholding full attestation guarantees.

Privatemode SDK

We're introducing a developer-friendly SDK designed to simplify the integration of Privatemode into your products. This SDK will minimize implementation effort and accelerate the time-to-market for your confidential AI applications.

Throughput and performance

We're continuously optimizing our infrastructure to reduce inference latency and improve throughput. High performance is a core product priority, ensuring that you get the fastest possible responses without sacrificing security or privacy.

Next-gen open source models

Support is coming for the latest generation of NVIDIA GPUs (Blackwell). This update will enable state-of-the-art LLMs (e.g., Kimi-K2.5, GLM-4.7), delivering higher performance while maintaining full data confidentiality.

Document processing and OCR

We're building secure document processing capabilities to enable the analysis and extraction of insights from a wide range of document formats using AI. This feature ensures that sensitive data remains fully protected throughout the entire workflow.