Insights
Blog & Insights
News, background, and honest stories from the Personarium workshop.
Images and voices with your own keys: how media generation works in Personarium
No second subscription for images or read-aloud: just like text chat, image generation, TTS, and optional video run on your own API key — with no markup.
What Does AI Chat Really Cost? The Cost Calculator Explained
Personarium doesn't mark up model costs. What you actually pay depends on the model you pick — our cost calculator shows it live instead of made-up numbers.
Why a membership instead of an AI subscription? The €4.99 model, honestly explained
€4.99 a month for AI chat sounds too cheap to be true. It is, because it isn't one: the membership covers something different from your AI usage, and that's on purpose.
How the Hub Is Moderated: Upload Checks, Review, Reporting
The public Hub is strictly SFW — enforced technically, not just promised. A look inside the moderation pipeline behind every published card.
Your character, your device: the architecture decision against plaintext server sync
A convenient cloud sync for your chats was on the table. We deliberately decided against it — and explain why that's more than a missing feature.
Encryption at Personarium: the local vault, and why we think quantum-safe
The vault encrypts selected characters and chats with your own password — purely symmetric, 256-bit, and built from day one for a world with quantum computers.
BYOK Explained: Why Personarium Runs on Your Own Key
Bring your own key: Personarium doesn't run its own AI and doesn't mark up model costs. What that means for you.
Sternwarte: Why Personarium Looks Like a Night Sky
The design story behind Personarium: midnight blue, a single ice-blue accent, and calm instead of neon.
Why Your Chats Never Leave Your Computer
An honest look at Personarium's BYOK architecture: your own keys, your own machine, no server reading along.