Video Generation

Today's Stories

1. 1

   xAI updates Grok to create videos from still photos

   Elon Musk's xAI company released Grok Imagine 1.5, which can turn any still image into a short video clip at 720p resolution based on written instructions. Users can combine multiple clips to create longer video scenes. The feature is available through xAI's platform as a preview version.

   Content creators and marketers can now animate their photos without video editing skills, potentially making video content creation more accessible for social media and presentations.

2. 2

   Google fixes billing bugs that overcharged Gemini AI video users

   Google resolved several issues with its Gemini AI app where users were incorrectly charged for failed video generation requests and where just one or two video creations would exhaust their entire monthly quota. The company also doubled the number of videos that paid Ultra subscribers can generate per month.

   Paying Gemini users will get better value from their subscriptions and won't lose credits when the AI fails to generate videos properly.

3. 3

   Video AI startup Luma launches open robotics training lab

   Luma AI, known for its video generation technology, is opening a robotics laboratory where anyone can train robots using the company's software. This represents the startup's expansion beyond creating AI-generated videos into physical robot applications.

   Small businesses and researchers may soon have easier access to robot training tools that were previously only available to large tech companies.

4. 4

   Users migrate to new video AI tools after Sora shutdown

   Following OpenAI's Sora video generator going offline a month ago, users have switched to alternatives like Kling, Runway, and CapCut Video Studio. Many creators report that newer tools like Seedance 2.0 provide similar quality with simpler workflows.

   Video creators have multiple reliable alternatives to choose from, ensuring continued access to AI video generation despite major platform changes.

5. 5

   Open-source avatar video generator LongCat 1.5 launched

   Chinese company Meituan released LongCat-Video-Avatar 1.5, an open-source tool that creates realistic talking human videos from audio input. The upgraded version uses Whisper-Large technology for better lip synchronization and can generate longer, more stable videos.

   Developers and content creators can now access professional-quality avatar video generation without paying subscription fees to major tech companies.