OpenCauldron Cloud · Invite-only beta

The open source AI creative studio

Generate images, video, and audio across Flux, Imagen, Ideogram, Recraft, and every major provider — with prompts, assets, brands, and team workflows in one workspace.

We'll only email you when your beta invite is ready. No spam, ever.

No setup

Sign in with Google, get a workspace, start generating in seconds.

Free during beta

Free credits while we're in invite-only beta. Paid tiers later.

Bring your own keys

Use your own OpenAI / Replicate / model provider keys — or use ours.

Same open code

Identical to the source-available studio you can self-host. No lock-in.

Join the community

10+

AI Models

100%

Open Source

BYO

API Keys

Free

During Beta

Brew media with any model

One interface for every AI model. Generate, manage, and track — all in one place.

9:41

Cauldron

Choose a model

Image Models

5 available
Gemini

Gemini

Photorealistic, high quality

$0.04
Imagen 4 Flash Ultra Fast ~8s
Grok

Grok

Fast generation, creative styles

$0.02
~5s
Flux

Flux

High quality, fast

$0.04
~6s
Ideogram

Ideogram

Best-in-class text rendering

$0.06
Create
Library
Brands
Profile

Choose from 5+ AI models — each with different strengths and pricing.

Everything you need to generate at scale

Not just another image generator. OpenCauldron is a full studio for teams that take their creative output seriously.

Multi-Model Generation

Flux Pro, Imagen 4, Ideogram 3, Recraft V3, and more. One interface, many engines — pick the best model for each job.

Bring Your Own Keys

Use your own API keys and pay providers directly. No markup, no middleman. You control your costs.

Brand Management

Tag generations by brand, organize assets, and keep creative output consistent across teams.

Usage & Cost Tracking

Monitor API costs per user, per model, per brand. Know exactly where your budget goes.

Team Collaboration

Shared gallery, leaderboards, and user profiles. See what your team is creating.

Self-Host Everything

Your database, your storage, your auth provider. Ships with sensible defaults, but nothing is locked in.

Full cost visibility

Know exactly where your budget goes

Set per-user daily generation limits, see exactly what each model costs, and track spend in real time. No more billing surprises.

Per-user daily limits

Cap generation counts per role. Admins, editors, and viewers each get their own ceiling.

Per-model cost breakdown

See spend by Flux Pro, Imagen, Sora, and every model your team touches.

Real-time spend alerts

Get notified when users or teams approach their limits. Stay ahead of overages.

USAGE OVERVIEW

Today

47 / 200

Today Cost

$12.40

+$3.20 since noon

This Week

312

This Month

$847.20

of $1,500 budget

USAGE BY MODEL

Flux Pro
156 gens $420.00
Imagen 4
89 gens $245.00
Sora 2 Pro
34 gens $142.00
DALL-E 3
33 gens $40.20

TEAM MEMBERS

AL

Adam L.

admin
$312.40
SK

Sarah K.

editor
$228.10
AT

Alex T.

viewer
$67.50
Organized creative library

Every asset, searchable and tagged

Every generation is automatically catalogued with its prompt, model, parameters, and creator. Organize by brand, filter by anything, and never lose track of what your team has created.

Brand-scoped galleries

Each brand gets its own asset library. Keep client work separated and organized.

Prompt & parameter history

Every generation stores its full recipe. Reproduce or iterate on any result.

Filter by model, date, creator

Powerful search across your entire library. Find any asset in seconds.

All Media
All Models
Search prompts...
Last 30 days
Flux Pro Acme Co
Imagen 4 Acme Co
Sora 2 Widget Inc
DALL-E 3 Acme Co
Ideogram Widget Inc
Flux Pro Acme Co

ASSET DETAILS

Prompt

A glowing cauldron made of obsidian stone, bubbling with iridescent purple liquid, ancient runes floating upward...

Flux Pro Black Forest Labs Acme Co

Aspect Ratio

16:9

Dimensions

1920 x 1080

Steps

28

Adam L. 3 hours ago
Gamification built in

Level up your creative team

OpenCauldron turns generation into progression. Every creation earns XP, unlocks achievements, and moves you up the leaderboard. Video generation unlocks at Level 3.

Earn XP for every generation

Image and video generations award XP. Pro models give more.

Unlock feats and achievements

Hit milestones, build streaks, explore models — earn badges along the way.

Climb the leaderboard

Monthly generation rankings plus all-time XP and feat leaderboards.

AL

Adam L.

admin · 892 generations · joined Mar 2026

4

Enchanter

520 / 800 XP to Level 5

520 XP
Lv. 4 Enchanter Lv. 5 Warlock

Feats Earned

🧪

First Brew

+10 XP

🏆

Centaur

+50 XP

🔥

Kindling

+25 XP

🧭

Ranger

+25 XP

🎬

Illusionist

+20 XP

🔒

Hydra

1,000 creates

Leaderboard — Highest XP

1
Adam L. 2,480 XP
2
Sarah K. 1,920 XP
3
Alex T. 1,150 XP

Powered by the best AI models

One interface, many engines. Choose the right model for each job.

Recraft

Recraft

V320BV4V4 Pro
Ideogram

Ideogram

3
Flux

Flux

1.1 ProDevKontextKlein
Grok

Grok

ImaginePro
OpenAI

OpenAI

2.01.51.0Mini
Gemini

Gemini

Imagen 4FlashLiteUltraFast
Stability AI

Stability AI

SDXL
Request a provider

Built for every creative team

Whether you're a solo creator or a 50-person agency, OpenCauldron adapts to how you work.

Individuals

One interface for every model. Stop juggling subscriptions and browser tabs. Generate, compare, and download — all in one place.

  • Compare outputs across models
  • Bring your own API keys
  • Track your spend in real time

Brands

Keep creative output on-brand. Organize assets by brand, set style defaults, and give your marketing team guardrails that actually work.

  • Brand-scoped generation
  • Style presets and prompt templates
  • Per-brand cost visibility

Agencies

Manage multiple clients from one instance. Per-brand separation, team access controls, and spend tracking across every client.

  • Multi-client brand management
  • Role-based team access
  • Per-user and per-client billing

Contributors

Help shape the open studio. Browse the codebase, fix bugs, add models, ship features.

  • Source-available license
  • Good first issues to tackle
  • Active Discord community

How we handle your data

Four commitments we can make today — verifiable in the source, not in a marketing PDF.

We don't train on your work

Your prompts and generations stay yours. We don't train models on them or share them with third parties. When you bring your own provider keys, that provider's policy applies on their side.

Your keys, your spend

Bring your own OpenAI, Replicate, or model-provider keys. Requests pass through — providers bill you, no markup, no key escrow.

Export and exit anytime

Workspaces, prompts, and assets export to the same schema as self-host. Walk away with your data intact — no lock-in.

Same code, in the open

Cloud runs the source-available studio. Audit it, fork it, or self-host the exact same build whenever you want.

Frequently asked questions

Pricing

Is OpenCauldron really free?

Depends on how you run it. Self-host is free and source-available under the Sustainable Use License — you only pay providers like Google or Replicate directly for AI model API usage with your own keys. Cloud is free during invite-only beta; paid tiers will come later, and you can always bring your own provider keys to keep costs predictable.

Setup

What API keys do I need?

You only need keys for the providers you want to use. For example, a Google AI API key gives you access to Imagen 4 and Veo. A Replicate token unlocks Flux Pro. You can start with just one provider and add more as needed.

Custom models

Can I add custom models?

Yes. OpenCauldron has a provider registry that makes it straightforward to add new models. You can fork the repo and add any model that has an API. We also welcome PRs to add new providers to the main project.

Billing

How is billing handled?

On self-host, OpenCauldron doesn't handle billing — your API keys are charged directly by each provider at their standard rates. On Cloud, you'll see provider costs the same way plus a monthly subscription for the hosted convenience (free during beta). Either way, the built-in usage dashboard breaks down every generation's cost by user, model, and brand.

Privacy

Where is my data stored?

On self-host, everything stays on your infrastructure — prompts and metadata in your PostgreSQL database, generated assets in your Cloudflare R2 bucket (or any S3-compatible storage). The self-hosted build never phones home or collects telemetry. On Cloud, your data lives in our managed PostgreSQL and R2 with per-workspace isolation — same code as self-host, just running for you.

Hosting

Can I use it without self-hosting?

Yes — OpenCauldron Cloud is the hosted version of the studio. Sign in with Google, get a workspace, start generating — no Docker, no setup. It's in invite-only beta right now; request a spot from the Cloud toggle at the top of the page. Prefer to run your own? Docker takes a few minutes — see the installation docs.

License

What license is OpenCauldron under?

OpenCauldron is released under the Sustainable Use License — the same license n8n uses. It's source-available: free for any internal business, personal, or non-commercial use, free to fork, modify, and contribute back. The one thing it prevents is third parties selling it as a competing commercial service. This keeps the project sustainable while staying open to almost everyone who wants to use or improve it.

Still have questions?

Browse the docs, ask in Discord, or open a GitHub discussion — the team and community are around.

Join the community

OpenCauldron is built in the open. Star us on GitHub, join the Discord, or dive into the code and contribute.