Image and video decision
Dezgo
A technical user can build a narrow self-hosted replacement (eg. text→image or a single text→video model) using open-source repos, but reproducing Dezgo's multi-model hosted catalog, scale, and polished UX is substantial and operationally costly.
Visit website↗Open-source builds that already do this
Every project below is open source and already does this job today. Fork one, self-host it, or take the parts you need — the build prompt further down assumes an empty file, and this is the shortcut past that. Licences differ; check the one on each card before you ship.
What a replacement has to do
- Accept a text prompt (and optional image/seed), run a generative model, store and serve resulting image/video, let user preview and download
What it still won’t have
- Hosted multi-model catalog and model-switching convenience
- Scale and reliability of a production service
- Polished web UI and user onboarding flows
- Free hosted usage for casual users
What remains hard
- Product polish and ongoing maintenance
First-year cost
No published price
Dezgo does not publish a price we could read, so there is nothing to compare against. What building costs is below.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build a minimal hosted AI image/video generation service using: Next.js (React) frontend, FastAPI backend, Postgres for jobs, Redis for queue, MinIO or S3 for storage, and PyTorch-serving workers running open-source models from open-mmlab/CogVideo. Scope: accept text prompts and optional seed images, enqueue jobs, run a text-to-image or text-to-video model to produce outputs (images or 3–15s video), generate thumbnails, persist outputs to object storage, provide job status API and a web UI to preview and download. Out of scope: training new models, multi-tenant billing, large model catalog, advanced editor features. Include logging, basic auth, retries for worker failures, end-to-end tests for API and a smoke test for a generation job, and Docker Compose + deployment instructions for one GPU instance.
How we checked
How the score was reached
- Partly verdict base52
- An open-source build was found+5
- 5 cited sources+3
- 3/3 assessment runs agreed+4
- Evidence score64
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time — so the same evidence always produces the same number.
How scoring works →Cited sources · 5
Every page the run actually retrieved.
- official productDezgo homepage
- official pricingDezgo pricing (not found)
- official docsDezgo features (not found)
- open sourcehacksider/Deep-Live-Cam
- open sourceAaronFeng753/Waifu2x-Extension-GUI
Integrity checks
What held up, and what did not.
