Elon Musk ’s AI startup xAI is all set to rollout a new text-to-video generation feature for its chatbot Grok this October. The new capability, dubbed “Imagine”, will allow users to generate short videos from text prompts, marking a major leap in multimodal AI . The feature will be powered by Aurora, xAI’s proprietary video generation model, and is expected to be integrated directly into Grok’s interface on X (formerly Twitter). Early demos suggest that users will be able to input natural language descriptions and receive AI-generated video clips within seconds.
The upcoming feature was reported by DogeDesigner and the post was later shared by Elon Musk. Unlike existing tools that rely on static image sequences or limited animation, Aurora is designed to produce fluid, context-aware video with realistic motion and scene transitions. The model has reportedly been trained on a diverse dataset of public video content and synthetic simulations, allowing it to handle a wide range of prompts—from cinematic storytelling to educational visualizations.
The launch of “Imagine” comes as xAI ramps up its competition with rivals like OpenAI’s Sora and Google DeepMind’s video models. Musk has previously stated that Grok will evolve into a fully multimodal assistant, capable of understanding and generating text, images, audio, and video.
The feature is expected to be available to Grok Pro subscribers first, with broader rollout planned later in the year.
The upcoming feature was reported by DogeDesigner and the post was later shared by Elon Musk. Unlike existing tools that rely on static image sequences or limited animation, Aurora is designed to produce fluid, context-aware video with realistic motion and scene transitions. The model has reportedly been trained on a diverse dataset of public video content and synthetic simulations, allowing it to handle a wide range of prompts—from cinematic storytelling to educational visualizations.
The launch of “Imagine” comes as xAI ramps up its competition with rivals like OpenAI’s Sora and Google DeepMind’s video models. Musk has previously stated that Grok will evolve into a fully multimodal assistant, capable of understanding and generating text, images, audio, and video.
The feature is expected to be available to Grok Pro subscribers first, with broader rollout planned later in the year.
You may also like
IPO-Bound Lenskart In The Black, Posts INR 297 Cr Profit In FY25
Ex-darts world champion pulls out of tournaments 'to get his eyes lasered'
Mumbai News: BMC Chief Bhushan Gagrani Reviews Dust Pollution Measures At Dadar's Shivaji Park, Proposes Mahim Fort Tourism Push
Maharashtra: New 104-km Expressway To Link Vadhvan Port With Samruddhi Mahamarg via Igatpuri
Canada's arms shipments to Israel continue despite official freeze, shows data; records reveal ammo and parts sent
