For the first time, users can enter an AI-generated world as themselves and explore it alongside others in real time, with no session limits.

PixVerse today announced a major evolution of R1, its real-time world model. Launched in January 2026 as the world's first real-time world model, R1 redefined video from a fixed output into a continuous, interactive stream — a world that responds instantly to user input. This update takes that world further.

This press release features multimedia. View the full release here: https://www.businesswire.com/news/home/20260401592938/en/

R1 now supports multi-user participation through a continuous shared livestream. The result is a shared digital world that evolves not from a single user's input, but from the aggregate input of everyone in it.

Users can now create a digital representation of themselves, place it inside the newly generated world, and share that experience with others in real time, with no session limits.

An Infinite, Interactive World

R1 generates a continuous, interactive visual environment from a text prompt. Unlike conventional video generation which produces a fixed clip, R1 creates an infinite stream that responds to user input as it unfolds, creating a world that can be shaped, directed, and explored. This remains the foundation of the R1 experience.

Step Inside this Digital World as Yourself

Users can now create a personalized digital avatar, accessible through a new Avatars tab. By uploading one to three photos of their front, side, and back views, users generate a virtual character that represents them inside the world. Avatars respond dynamically to interaction including movement, positioning, and animation, giving users a personal identity within the generated environment.

From Solo to Shared, Now Without Limits

Previously, R1 offered individual sessions of up to five minutes. This update removes that boundary entirely. The shared world runs continuously, with no session limit, evolving as long as users are in it.

R1 now supports multi-user participation through a continuous shared livestream. Users submit prompts into a common live feed, each prompt visible to all participants as it scrolls across the screen. When the AI picks up a prompt, it is instantly realized as video within the shared environment, accompanied by real-time visual effects. A live chat layer enables users to interact alongside the evolving stream, creating a dynamic feedback loop between participants and generation.

The result is a shared digital world that evolves not from a single user's input, but from the aggregate input of everyone in it.

The R1 Real-Time World Model

Three core components make continuous, multi-user video generation possible at scale:

• At its core, an Omni Native Multimodal Foundation Model processes text, image, audio, and video as a single unified stream rather than handling each separately, enabling coherent real-time generation across all modalities.
• A Consistency-aware Autoregressive Framework ensures that generated sequences remain temporally coherent over time, enabling continuous video without resets.
• An Instantaneous Response Engine reduces processing steps, enabling real-time 1080p generation with millisecond-level latency.

PixVerse has consistently shipped model updates every two to three months since its founding. This R1 expansion, arriving less than three months after R1, continues that commitment. This update is one of several releases PixVerse has made this week, following V6 on March 30 and Team Plan, Mini Apps, and PixVerse Skills on March 31.

Looking ahead, PixVerse is focused on improving the experience for users experimenting with this new mode of interactive video generation. The team is also exploring a broader range of use cases, including more immersive applications in entertainment and gaming, as the technology matures.

R1 is available free of charge to all PixVerse users for a limited time at realtime.pixverse.ai.

ABOUT PIXVERSE

PixVerse is a global AI video generation platform trusted by over 100 million creators and enterprises across 175 countries. PixVerse's latest model, V6, builds on its foundation of fluid motion and cinematic quality, adding precision camera control, expressive character performance, and one-click commercial output. In January 2026, PixVerse launched R1, the world's first real-time world model, transforming video into an infinite, continuous, and interactive stream. With teams distributed across Asia and the US, PixVerse was founded in 2023 with a commitment to making video the universal language of human expression. In March 2026, PixVerse closed its Series C funding round, achieving unicorn status. For more information, visit pixverse.ai.

View source version on businesswire.com: https://www.businesswire.com/news/home/20260401592938/en/

Users can now create a digital representation of themselves, place it inside the newly generated world, and share that experience with others in real time, with no session limits.