Video
Image to Video
Start from one still frame and let a video model add movement, camera motion and atmosphere while your image keeps the look consistent.
Try NowReference to Video
Upload reference images to lock in subject and style, add a prompt for motion and mood, and generate video that stays visually consistent.
More models on Hushgen
One credit balance works with all of them.
Video models
Models are provided by their developers through licensed API services. Hushgen is an independent platform; model names belong to their respective owners. Disclaimer
AI Tools
More ways to create
Switch to another tool any time. Each one lets you pick from several AI models and uses the same credits.
Text to VideoImage to VideoReference to VideoVideo ExtendText to ImageImage Edit
Video
Text to Video
Describe a scene in a sentence or a short script and get back a clip you can refine shot by shot.
Try NowVideo
Reference to Video
Upload reference images or clips to pin down a character, product or visual style before the model starts generating.
Try NowVideo
Video Extend
Pick up where a clip ends and add the next few seconds with the same subject, framing and mood.
Try Now
ImageText to Image
Write a prompt and get thumbnails, product shots, concept art or posters, with Seedream and other image models to choose from.
Try Now
ImageImage Edit
Change a background, remove an object or restyle a photo by telling Hushgen what to change in plain words.
Try NowGenerate Consistent Video from References
Upload images that show the subject, style or composition you want, then describe the motion. Hushgen uses them to keep every shot visually on track.
What Reference-Guided Video Gives You
More control over how your subject and style carry from frame to frame.
Image References
Upload 1-3 reference images to guide the visual style, color palette, subject, and composition of the generated video.
Image-Only References
Veo 3.1 Reference to Video uses image references only. Video uploads are not supported by this model.
Reference Image Set
Use the first reference image as the strongest subject cue, then add up to two more optional images for style and composition context.
Prompt Control
Add a text prompt to describe the video style, motion, and mood. Combine with references for precise creative control.
Prompt Only
Describe the action, framing, and mood directly in your prompt. Prompt expansion is not submitted for Veo 3.1 Reference to Video.
Full-Resolution Output
Download your generated video in full resolution — ready for social media, presentations, or any professional use.
From References to Clip
Upload references, add a prompt, generate.
1
Upload Your References
Upload 1-3 reference images that represent the subject, visual style, and composition you want in the output.
2
Add a Prompt
Describe the video style, motion, and mood in plain text. The Veo 3.1 request is submitted with your prompt and reference images.
3
Download the Video
Your style-consistent video is usually ready in a few minutes. Download it in full resolution and use it anywhere.
Where Reference-Guided Video Helps
Any project where the same face, product or look has to stay recognizable.
Brand Video Production
Generate videos that match your brand's visual identity by using existing brand assets as style references.
Style Transfer
Apply the visual style of your reference images to new video content — consistent color grading, lighting, and aesthetic.
Character Animation
Use character or product images as subject references and guide the intended movement through your prompt.
Creative Exploration
Combine multiple reference styles to explore new visual directions and generate unique, unexpected results.
Social Media Content
Create on-brand video content for Instagram, TikTok, and YouTube using your existing visual assets as guides.
Film & Production
Generate style-consistent video sequences for pre-visualization, mood boards, and concept development.
Choosing References That Work
How to pick and combine images for steadier, cleaner results.
Use References with Consistent Lighting
Reference images with similar lighting conditions produce more cohesive output. Mixing bright outdoor and dark indoor references can create inconsistent results.
Choose References That Share a Visual Style
The AI blends the style of all your references. For the most consistent output, use references that share a similar color palette, contrast level, and visual aesthetic.
Use Focused Reference Images
Use clear, focused images with one main subject or style direction. Describe motion directly in the prompt.
Use the Reference Image Set for Subject Control
Place the most important subject or style in Image 1. Additional images should support the same visual direction rather than introduce conflicting cues.
Combine Up to Three Images
Use up to three aligned images to guide subject, style, and composition. Use the prompt to describe motion and camera behavior.
Write Specific Prompts for Complex Styles
When working with complex or nuanced reference styles, write the key style, motion, and camera details directly in your prompt.
Who Uses Reference to Video?
Teams and creators who need their own visuals to lead the generation.
Brand & Marketing Teams
Generate videos that match your brand's visual identity by using existing brand assets — photography, style guides, and campaign videos — as references.
Animators & Motion Designers
Use image references plus detailed prompts to guide motion style and character animation concepts.
Film & Video Producers
Generate style-consistent pre-visualization footage, mood board videos, and concept sequences using reference materials from the production.
Social Media Agencies
Create on-brand video content for clients by using their existing visual assets as style references — consistent output at scale.
Game Developers
Generate cinematic sequences and environment videos that match the visual style of your game using concept art and image references as guides.
Creative Directors
Explore and validate visual directions quickly by generating reference-guided video concepts before committing to full production.
Why Use References on Hushgen
Consistency, control and a clear view of what each generation costs.
Style Consistency
Use your own reference materials to ensure the generated video matches your exact visual style and brand identity.
Fast Generation
Most videos finish within a few minutes, depending on the model, length and resolution. Paid plans get priority in the generation queue.
Privacy First
Your prompts and files are processed securely and sent only to the AI model provider that generates your result. We never sell your data.
Pay Only for What You Use
Credit-based pricing means you only pay for the videos you actually generate — no monthly minimums.
Reference to Video Questions
How references shape AI video, answered.
What types of reference files can I upload?
For Veo 3.1 Reference to Video, upload 1-3 reference images (PNG, JPG, WEBP). This model does not support reference video uploads.
What is Image 1 used for?
Image 1 is the strongest subject cue for the generated video. Additional images can reinforce the same subject, style, or composition.
Do I need to upload videos?
No. Veo 3.1 Reference to Video requires images only: minimum 1, maximum 3 reference images. Video uploads are not supported on this model.
How does the prompt work with references?
The prompt describes the video style, motion, and mood in plain text. It works together with your references — the references guide the visual style while the prompt guides the content and action.
What is prompt expansion?
Prompt expansion is not submitted for Veo 3.1 Reference to Video. Write the important action, style, camera, and mood details directly in your prompt.
How long will the generated video be?
Video length depends on the model and settings selected. Available duration options are shown before you generate.
How many credits does a generation cost?
Credit cost depends on the model, resolution, and duration selected. The required credits are shown before you generate, so there are no surprises.
Is my data private?
Yes. Your prompts and files are processed securely and sent only to the AI model provider that generates your result. We do not sell your data or use it to train our own models. See our Privacy Policy for details.
Can I use generated videos commercially?
Paid plans include commercial usage rights and watermark-free output for everything you generate. Free plan outputs include a watermark. See our Terms of Service for details.
How many reference files should I upload for best results?
For most use cases, 1–3 focused reference images produce the best balance of style consistency and generation quality. Too many conflicting references can dilute the style signal.
Can I use screenshots from a film or TV show as references?
Technically yes, but be mindful of copyright. Using copyrighted material as a reference to generate commercial content may raise legal issues. For commercial projects, use your own assets or royalty-free reference materials.
What is the difference between reference-to-video and text-to-video?
Text-to-video generates content purely from a written description, giving the AI full creative freedom. Reference-to-video uses your uploaded images to constrain the visual style and subject direction, producing output that is consistent with your specific reference materials.