Montagea

What is Montagea?

Montagea is GitHub for AI video: an AI-first platform where you write a story, split it into scenes, and render every shot with whichever model fits it best. You can make the whole film yourself, or open scenes up to a community of creators who render their own takes. You review them the way a developer reviews pull requests.

Every video on Montagea is generated on the platform, and every accepted scene is permanently credited to the person who made it.

Who it's for

• Writers: bring a logline, a poem, or a full screenplay. AI polishes it and turns it into a shot list, and a crew can render the rest.
• Directors and creators: pick the engine for each scene, keep characters consistent, and assemble the final cut in one workspace instead of 47 browser tabs.
• Contributors: browse works that are casting, claim an open scene, render your take, and earn credits (or real money on funded open calls) when it's accepted.
• Marketers, musicians and small studios: produce shorts, music videos, ads and prototypes without switching between five subscriptions.
• Anyone curious about AI video: every model has its own playground page, so you can compare them before you commit to one.

What you can do

Generate video with 20+ frontier models

One account and one credit balance cover 10 video engines, including Google Veo 3.1, Gemini Omni Flash, Seedance 2.5, MiniMax Hailuo H3, Wan 3.0, Grok Imagine, Runway Gen-4.5 and more.

• Text-to-video, image-to-video, and first/last-frame interpolation
• Reference-to-video: combine reference images, video clips and audio to keep characters and style consistent
• Multi-shot storyboards that generate a coherent sequence in one pass
• Video edit and extend on models that support it
• Output from 720p up to 1080p and 4K, with native audio on models that support it
• AI avatars and lip-sync: turn a photo and an audio track into a talking or singing video, or re-sync the lips in existing footage

Create images, voices and music in the same place

• Images: Qwen-Image, Nano Banana Pro, GPT Image, Seedream, FLUX.2, Z-Image, Wan Image and more. Use them for character sheets, concept art, and first frames that feed straight into video.
• Voice: Gemini TTS and Fish Audio, with dozens of voices in 20+ languages.
• Music: Google Lyria 3, MiniMax Music and Mureka, for songs with lyrics or background scores.
• Writing: 20+ LLMs (GPT, Claude, Gemini, DeepSeek, Qwen and others) to draft, polish, translate and continue scripts.

Reverse-engineer what you love

Upload an image to get a detailed prompt for it, or upload a video to break it into a shot-by-shot storyboard with a prompt and keyframe for each shot. It's a quick way to learn how a shot was built, or to rebuild it in your own style.

Make films together

1. Script: write your story or generate one with AI.
2. Plan: AI splits it into scenes, with framing, mood and reference frames.
3. Render: generate each scene with the engine that fits, yourself or through the community.
4. Merge: review submissions (accept, reject, or request a revision). Accepted scenes are assembled into a final cut automatically. You can upload your own edit at any time and roll back to any earlier version.

Playground and multiplayer rooms

A chat-style playground for every modality. Invite friends into a shared room, generate side by side, and react to each other's results in real time.

Live channels

Interactive AI video livestreams: viewers prompt the next clip, and it airs for everyone in the channel.

Discover and share

An explore feed, trending works, creator profiles, follows, comments and a leaderboard. Published works get their own watch page.

Why Montagea

• The right engine for every shot. No single model is best at everything. Veo is strong at cinematic action with sound, Seedance at realistic people and long takes, MiniMax H3 at mood, and Wan at stylized looks. Montagea lets you choose per scene, and new models usually arrive the week they launch.
• One bill instead of many subscriptions. Credits work across every video, image, voice, music and text model. You see the price before you generate, and failed generations are refunded automatically.
• Collaboration is built in. Scenes, claims, submissions, reviews and version history work together the way code collaboration does.
• Credit that sticks. Every accepted contribution is permanently attributed to its creator.
• Available in English, Chinese (简体中文) and Japanese (日本語).

Pricing

• Free: no card required. You get 100 credits at signup (500 with an invite code), plus daily free generations on selected models (MiniMax H3 720p video, Z-Image and Qwen-Image).
• Starter / Pro / Premium: monthly credits, access to every model, 1080p and 4K output, avatars and lip-sync, monthly allowances on self-hosted models, priority queue and more concurrent generations. Annual billing saves up to 12%.

Start free and upgrade at any time.