Blog AI
Generates long-form blog posts in a trained brand voice through research, outline, drafting and proofreading stages.
A staged pipeline: cited research, outline, section drafts, proofreading, optional SEO loop.
Blog AI generates long-form blog posts and books in a trained brand voice, then publishes them. It suits small teams who want a self-hosted, staged writing pipeline with citations and are comfortable running a web app and API.
What it is
The README says a post is researched with web search (up to 8 cited sources), outlined, written section by section with a brand voice summary, then proofread and humanised. Two opt-in stages add an SEO rewrite loop and a claim-by-claim fact check.
Who it's for
Suits technical founders and content teams who want diagnosable stages and their own LLM keys. It does not suit anyone who needs short-form ad copy or a polished hosted product: it is a small project.
Key features
Staged generation
Research, outline, per-section drafting, proofreading.
Cited research
Web search with up to 8 cited sources.
Brand voice summary
Sections are written with a trained brand voice.
SEO rewrite loop
Opt-in loop rewrites until a score threshold is met.
Fact check
Opt-in claim-by-claim verification stage.
Browser extension
Includes a browser extension alongside the web and API apps.
Up and running in 3 steps
Our honest verdict
What's great
- Each generation stage is a plain function with its own tests, so failures are diagnosable.
- Works with OpenAI, Anthropic or Gemini behind one interface.
- Cited research and an optional fact-check stage.
- Runs without a database in development, using in-memory storage.
Where it falls short
- AGPL-3.0: if you run a modified version as a service you must share your changes.
- Requires Bun 1.4.3, Python 3.12 and an LLM key; production uses Clerk, Neon and Stripe.
- Small, young project with a thin community compared with commercial tools.
- Aimed at long-form posts, not short ad or social copy.
Pricing & license
Free under AGPL-3.0. The README links a hosted version at blog-ai.vivancedata.com; no price is stated in the README. Self-hosting costs your own LLM API usage.
Open remote AI-training roles
767 open remote AI-training roles, updated November 14, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.
Alternatives
These reviewed AI tools do the same job instantly — from our scored directory.
Use it, or have us set it up
FAQ
Is Blog AI free to self-host?
Yes under AGPL-3.0, with the obligation to share source changes if you offer it as a network service. You pay for your own LLM API usage.
Does it cite sources?
The README says research uses web search with up to eight cited sources.
Do I need a database?
Not for development: without DATABASE_URL, history and brand profiles fall back to in-memory storage.
