Compare AI video models. Know the price. Create in one workspace.
MaxVideoAI is a multi-model AI video production platform for turning a brief into finished, reusable work. Explore real examples, compare current engines side by side, review model capabilities and pricing, prepare characters and shots with production tools, generate video, image, or audio, and keep outputs and references together in the MaxVideoAI Library.
Plan a video with MaxVideoAI · Compare current models · Connect MaxVideoAI to your assistant or automation
Plugin repository · Plugin releases
MaxVideoAI brings model choice, prompts, references, settings, live pricing, previews, results, and next-shot context into one production workspace. That workspace is the main product; the GitHub plugin extends the same production path to Claude, ChatGPT, Codex, OpenClaw, and a bounded self-hosted n8n workflow later in this page.
The current public production experience, captured from production on September 16, 2026.
Use MaxVideoAI to develop a single shot, animate an image, build a multi-shot sequence, create supporting images or audio, and continue from earlier work. The workspace keeps model selection, prompts, references, settings, price, previews, results, and the Library in one production path instead of scattering them across provider tabs.
Every public example opens into its prompt, settings, model, duration, and recorded render cost. Browse the AI video examples gallery, then recreate a useful starting point in your own workspace.
| Production goal | Start here |
|---|---|
| Find a visual direction | Preview varied examples and inspect the prompt and settings behind them. |
| Build with references | Choose a workflow that supports the image, video, or audio material you want to use. |
| Plan several shots | Keep one brief visible while comparing a consistent route with an intentional model mix. |
| Continue finished work | Recover outputs and reusable media from the MaxVideoAI Library. |
MaxVideoAI compares AI video engines by workflow, supported inputs, duration, resolution, audio, pricing, strengths, and a documented editorial score. The side-by-side hub is designed for an actual production decision: choose two engines, inspect the trade-offs, open the detailed matchup, and move to the model that fits the shot.
The broader model directory adds recommended starting points and routes for video, image, audio, and preparation work.
Model families such as Sora, Veo, Kling, Seedance, LTX, MiniMax, Wan, Happy Horse, Gemini Omni, and Pika are examples rather than a frozen catalogue. Use the live directory for current availability and pricing.
MaxVideoAI is pay as you go. The pricing surface compares common scenarios across video, image, audio, and production tools, while the workspace shows the live price for the selected request before launch. In the assistant workflow, a concrete request becomes an exact quote and one explicit approval authorizes one paid attempt.
- Choose the model, mode, duration, resolution, and supported extras.
- Compare the current AI video pricing.
- Review the live workspace price or assistant quote.
- Launch from the web app, or explicitly approve one prepared assistant attempt.
The public production toolbox brings repeatable preparation and finishing work into the same product path. Build a consistent character sheet, explore camera angles, refine or remove an image background, and finish a video without pretending that every useful step needs a new generation model.
Tool availability and price still come from the live product. The exact input, output, and billing boundary is shown before a tool runs.
Supported workflows can use image, video, or audio references for composition, subject, motion, timing, or sound direction. The selected model and mode determine the accepted reference roles and limits. Finished results and reusable assets remain in the connected MaxVideoAI Library, ready for the next shot or another production route.
Open the Library to browse saved media and continue where the production stopped.
MaxVideoAI exposes one remote MCP production service through five public paths: Claude, ChatGPT, Codex, OpenClaw, and a tested self-hosted n8n workflow. A connected assistant or automation can help shape the brief, compare current model options, calculate project budgets, prepare an exact quote, wait for explicit approval, and recover the accepted result in the same MaxVideoAI account.
Claude setup · ChatGPT setup · Codex setup · OpenClaw setup · n8n setup · MCP technical reference · Public plugin source
OpenClaw is tested with limits and has a listed ClawHub skill. n8n is tested on self-hosted n8n through a manual setup of the deterministic MCP Client workflow; n8n Cloud and MCP Client Tool agent invocation are not claimed. Cursor, GitHub Copilot, Gemini CLI, and Microsoft Copilot remain in preparation and do not yet have public setup paths.
Brief → compare models → budget shots → prepare exact quote → approve one attempt → recover result
The focused plugin repository carries tagged, checksum-backed releases. Use its release page for the current installable package; this product repository remains the authored source of the application and plugin bundle.
This repository contains the production application, its public marketing and comparison surfaces, provider orchestration, billing controls, media workflows, model registry, remote MCP server, and the distributable MaxVideoAI plugin package.
flowchart LR
B[Creative brief] --> W[Next.js workspace]
B --> A[Assistant, agent or automation]
A --> M[Remote MCP + OAuth]
M --> W
W --> C[Capability-driven model catalogue]
W --> P[Server-owned provider orchestration]
P --> J[Jobs, billing and refunds]
J --> L[Private media Library]
L --> W
| Layer | Technology and ownership |
|---|---|
| Web product | Next.js App Router, React, TypeScript, Tailwind CSS |
| Authentication | Supabase Auth |
| Application data | Neon Postgres |
| Media storage | Amazon S3 |
| Billing | Stripe and server-owned price/approval controls |
| Model execution | Capability-driven registry plus provider adapters owned by the server |
| Assistant integration | Remote Streamable HTTP MCP, OAuth, checked-in Claude/Codex manifests and skills, plus documented OpenClaw and n8n paths |
| Quality | Node test contracts, Playwright browser checks, lint, typecheck, exposure and asset gates |
| Repository area | Responsibility |
|---|---|
frontend/app |
Routes, layouts, metadata, APIs, and route orchestration |
frontend/components |
Shared product and marketing UI |
frontend/lib |
Browser-safe product, pricing, routing, and SEO logic |
frontend/server and frontend/src/server |
Server orchestration, providers, jobs, billing, and MCP |
frontend/config/model-registry.json |
Authored model identity and publication policy |
plugins/maxvideoai |
Public plugin package, skills, host guides, examples, and release assets |
neon/migrations |
Application database migrations |
Read the project structure, model registry workflow, MCP mode coverage, and private reference-import boundaries before changing those contracts.
The contributor path uses Node.js 22 and the repository-pinned pnpm version:
pnpm install
cp frontend/.env.local.example frontend/.env.local
pnpm devFocused validation for public GitHub and plugin work:
pnpm github:content:check
pnpm github:assets:check
pnpm github:score -- --require-after
npm run lint:exposure
git diff --checkRead local development for the frontend, contract server, Docker, migrations, model onboarding, scheduled jobs, and validation commands. The environment reference documents variables, health endpoints, and the Supabase/Neon/S3 ownership boundary.
Before an architecture or model-policy change, start with AGENTS.md, the LLM working guide, and the closest area-specific instructions.
Small, reviewable contributions are welcome when they preserve public URLs, the authored model registry, server/client boundaries, private-media handling, and the evidence contracts that keep public claims accurate.
| Need | Start here |
|---|---|
| Propose or implement a change | CONTRIBUTING.md |
| Report a vulnerability privately | Email security@maxvideoai.com; do not open a public issue. |
| Ask for product or repository help | Contact MaxVideoAI |
| Review public/private boundaries | docs/public-vs-private.md |
The repository uses the Business Source License 1.1, with its terms and change date defined in the license file. Commercial deployments require a separate licence; see the dual-license guide, email licensing@maxvideo.ai, and review NOTICE.
Last reviewed: 2026-09-16.