Step 1
Brief the creative outcome
Start with the end in mind: a product launch clip, a character continuity test, a campaign variant, or a storyboard beat. Write the scene intent the way you would brief agent Mel — subject, mood, camera, and audio.
What is Melius
The creative canvas for agents — generate, iterate, and compose AI media in one workflow
Melius is a node-based creative canvas where prompts, images, video clips, and audio connect as nodes on an infinite canvas. The Mel agent routes each step to the right model, and you can steer every prompt until the output lands exactly as you imagined.
On this page you can run the same core production moves immediately — text-to-video, image-to-video, reference-led consistency, and native audio — from a streamlined studio built for creative directors, agencies, and production teams.
How to plan a Melius creative workflow
Brief the idea, pick the right input mode, steer the motion, and generate — the same creative director loop Melius is built around.
The quality of the brief drives the quality of the output. Structure your scene intent before fine-tuning visual details.
Step 2
Choose the generation mode
Pick text-to-video when the idea lives in words, image-to-video when a still frame already sets composition, or reference-to-video when identity, wardrobe, or product details must stay consistent across shots.
Step 3
Steer motion and audio together
Describe camera movement, pacing, scene energy, and audio mood in one prompt. Melius treats sound and movement as part of the brief, not a post step — so lead with how the shot moves and what it sounds like.
Step 4
Generate and iterate
Run the clip, review continuity and lip-sync, then refine the prompt or swap references. When the output lands, download in 1080p for client review, campaign delivery, or social publishing.
More AI video generators
Post-generation polish and planning
Key features of Melius AI
From multi-model access to reference-led consistency — every capability is built for real creative production.
Text, image, and reference workflows, native audio, 1080p delivery, and multi-model switching are designed to work together from brief to render — the same production stack Melius users rely on for advertising, e-commerce, and film.
01
One subscription, every image and video model
Melius routes prompts across dozens of leading AI models — Seedance, Veo, Sora, Kling, Flux, GPT Image, Nano Banana, and more. Switch models without leaving your creative workflow.
02
Reference-led consistency
When a brief depends on stable characters, products, or environments, reference-to-video delivers stronger continuity than prompt-only generation — with support for up to nine reference images.
03
Native audio built into the brief
Audio is not an afterthought. Melius treats synchronized sound as part of generation, so you plan pacing, ambience, and dialogue together rather than stitching them on afterward.
04
1080p delivery path
Output at higher resolution for client previews, product stories, branded shorts, and polished creator cuts where draft quality would not be enough.
05
Image and video in one canvas
Connect image generation to video generation in a single pipeline. Start with a concept still, extend it into motion, and keep visual direction consistent from frame to clip.
06
Multilingual lip-sync for global content
Create talking-head scenes, product demos, and ad variants across multiple languages with lip-sync that stays aligned — essential for campaigns that ship in every market.
Generation modes in Melius
Three ways to start a video — text, image, or reference. Pick the input that matches your creative process.
Melius text-to-video
Use text-to-video when the shot is already clear in your head and you do not need a source image to establish framing or identity.
Best for scripted camera ideas and scene-first prompting
Ideal for concept drafts, ad scene exploration, and visual story beats
Pairs well with audio-aware prompting from the start
Melius image-to-video
Use image-to-video when a still frame already carries the right subject, product, pose, or composition and you want motion to extend that frame naturally.
Great for product stills, album covers, and portrait motion tests
Better when framing accuracy matters more than scene invention
Requires less prompting because the starting frame already does the visual work
Melius reference-to-video
Use reference-to-video when the job is continuity across multiple shots. Support for up to nine reference images gives stronger identity and art-direction control than a single frame.
Purpose-built for characters, product systems, branded environments, and style boards
Designed for briefs where one image is not enough to anchor the result
The clearest choice when you need consistency across details, wardrobe, or props
Best use cases for Melius
Melius delivers the most value when your project spans advertising, e-commerce, filmmaking, and brand production.
01
Advertising and campaign variants
Turn a product mockup or lifestyle still into motion-driven ad cuts. Generate parallel variants for every placement and aspect ratio from one approved creative direction.
02
E-commerce product storytelling
Transform pack shots and catalog stills into motion previews with deliberate camera language — the same e-commerce workflow Melius is built for.
03
Filmmaking and storyboard validation
Test whether reference frames hold continuity before committing to a full production pipeline. Validate beats and camera language at pitch stage.
04
Brand and design lookbooks
Pin visual direction with reference images, then generate on-brand motion and stills. Keep characters, props, and environments recognizable from shot to shot.
05
Multilingual ad localization
Produce the same scene for multiple markets with aligned lip-sync and audio — expanding one creative concept across regions without rebuilding the workflow.
06
Agency concept previews
Present clearer motion direction at pitch stage when 1080p output and structured prompting are more important than raw turnaround speed.
Melius AI – FAQ
Everything you need to know about Melius — capabilities, modes, models, audio, and how to generate on this page.
What is Melius?
Melius is a node-based creative canvas for AI image, video, audio, and text. Every prompt, asset, and generation is a node on an infinite canvas, connected by edges so one step feeds the next. The Mel agent routes prompts to the best model automatically, and external agents can connect via MCP.
What is a node-based AI canvas?
A node-based canvas represents creative work as a graph instead of a chat thread. Every step is a node with its own model, settings, and version history — a prompt, a generated image, a video, an audio track, or an uploaded reference. Edges pass output from one node into the next, so you can rewire, branch, and reuse working pipelines as templates.
Can I use the Melius generator on this page?
Yes. The generator on this page supports Melius text-to-video, image-to-video, and reference-to-video workflows, so you can move from brief straight into generation.
What AI models does Melius support?
Melius gives you access to leading AI models for image, video, audio, and text — including Seedance, Veo, Sora, Kling, Flux, GPT Image, Nano Banana, and ElevenLabs for audio, from providers such as OpenAI, Google, ByteDance, and Black Forest Labs. The Mel agent can pick the best model for your prompt, or you can choose one yourself.
Does Melius support audio?
Yes. Melius generates synchronized native audio as part of video output, so you can plan sound, dialogue, and pacing in the same brief rather than adding audio in post.
Is Melius free?
You can explore Melius workflows on this page with credits. Generation uses a credit-based model — you pay for what you create, with transparent per-task pricing shown before you run.
How does AI video generation work on Melius?
You start from a text prompt or an existing image, choose a video model or let the agent pick one, and Melius generates the clip. You can guide it with reference images, first and last frames, or input footage, then refine and download the result.
How should I prompt Melius?
Brief like a creative director: define subject, camera movement, pacing, scene energy, and audio mood first. Pick the generation mode that matches your starting material. If continuity matters, give each reference image one clear job instead of packing too many visual instructions into a single frame.
Get started
Start creating with Melius now
Melius combines multi-model video and image generation with native audio, 1080p output, and reference-led consistency — giving you professional-grade creative output from the input of your choice.