Changelog
Stay up to date with the latest changes in our product
PixVerse V6 and MiniMax H3 are now available
v1.3.0Create videos from text or images with PixVerse V6 and MiniMax H3, with flexible durations, quality options, and frame guidance.
2026/08/02
Two New Video Models
- PixVerse V6 is now available for text-to-video and image-to-video generation, with 1–15 second output, 360p–1080p quality, optional audio, and multi-clip support.
- MiniMax H3 is now available for text-to-video and image-to-video generation, with 4–15 second output, 2K video generation, and first/last-frame guidance.
More Ways to Create
- Start with a written idea or bring a still image to life with natural motion.
- Use PixVerse V6 to create short clips with optional sound and multi-clip storytelling.
- Use MiniMax H3 when you want higher-resolution output or more control over the opening and ending frames.
Workspace Improvements
- Added dedicated logos and quick access for both new models in the workspace.
- Improved the sidebar on shorter screens so every creation tool remains easy to reach.
- Model descriptions and settings are now available in all supported languages.
Gemini Omni video model is live
v1.2.0Voe Ai now supports Gemini Omni for multimodal video generation, with new marketing and workspace entry points.
2026/05/22
New Model
- Gemini Omni is now available for text-to-video, image-to-video, reference-to-video, and video editing workflows.
- The model supports text prompts, reference images, and optional source video input for multimodal creation and editing.
- Creators can choose 16:9 or 9:16 output with 720p, 1080p, and 4K resolution options.
Workspace Updates
- Added Gemini Omni to the workspace New Models section.
- Added Gemini Omni to the top marketing banner for today's model launches.
- Published a dedicated Gemini Omni landing page with capabilities, use cases, FAQ, and pricing modules.
Platform Improvements
- Added Gemini Omni model settings, credit estimates, and the KIE provider handler.
- Added upload validation for Gemini Omni image and video assets, plus dedicated credit estimates when a source video is included.
- Updated the sitemap, footer model links, and anonymous page cache rules for the new marketing page.
GPT Image 2 and HappyHorse 1.0
v1.1.0New image and video models are now available in the Voe Ai workspace.
2026/05/19
New Models
- GPT Image 2 is now available for text-to-image and image-to-image workflows.
- HappyHorse 1.0 is now available for text-to-video, image-to-video, reference-to-video, and video editing workflows.
Workspace Updates
- Added dedicated Reference to Video and Video Edit entries in the workspace navigation.
- Video editing now supports an input video plus optional image references.
- Reference-to-video now supports ordered reference images for HappyHorse.
- Model settings, file limits, accepted file types, and credit estimates have been wired into the composer.
Platform Improvements
- Added generation routing, pricing, and KIE provider handlers for the new models.
- Improved generation asset handling for image URLs and video URLs.
- Kept existing Veo 3.1 workflows scoped to text, image, and keyframe video generation.
Voe Ai v1.0.0
v1.0.0A simple overview of what you can do with Voe Ai
2025/10/28
What You Can Do
- Turn your ideas into short videos with a simple text prompt
- Guide the style with up to 3 reference images
- Control the beginning and ending using first/last frame keyframes
- Choose your video shape: landscape (16:9) or vertical (9:16)
- Add your own watermark text to protect your video
Using Voe Ai
- Sign in easily and start creating videos right away
- Use credits to generate videos; buy more credits or subscribe any time
- See your balance and history on the Credits page
- Track progress and download your finished videos from the dashboard
- Manage your plan and payments on the Billing page
Notes
- Reference-image mode works best with 16:9 videos
- Very large images may fail to upload; try smaller PNG/JPG/WebP files if that happens