Frequently asked questions
What is Fervyn?
A local AI roleplay companion that leads with image generation — the same character, the right scene, panel after panel — and keeps your story as a picture-book on your own disk. It runs on your computer, not in the cloud.
What does it cost?
The Lite installer is a free download. There's nothing to buy to use what you install today.
Does it run on my machine or in the cloud?
On your machine. First launch downloads the models for your hardware — roughly ~28 GB for the 8 GB config, ~32 GB at 12–16 GB, ~38 GB at 24 GB and up (runtime stack + models) — resumable if your connection drops. After that, the companion experience runs locally — see the installation guide for the full band-by-band picture.
How do updates work?
New builds are announced on the changelog (RSS available). Updating means running the new installer — your characters, chats, and picture-books live in your user data folder and survive reinstall. Nothing updates itself behind your back.
Is it really private?
The companion chat path makes zero outbound network calls, and there's a public CI test that proves it (the build fails if that ever breaks). Full detail: Privacy & provenance. (Optional features that are off by default may use the network; they're never on the companion chat path.)
Do I need an account or to upload ID?
No account, and no ID upload, to use what you install.
Can I keep / move my creations?
Yes. Export any saved chat as a self-contained bundle — prose, images, and a verifiable C2PA + watermark record on each panel. It's yours; there's no platform that can take it down.
What hardware do I need?
An NVIDIA GPU with 8 GB VRAM on Windows (RTX 20-series or newer recommended) — and 8 GB now runs a capable 12B writer, not a small one. More VRAM upgrades the configuration automatically: a 12 GB card (RTX 3060 12GB / 5070) adds the full-quality image engine, 16 GB adds no-wait speed and long-term memory, and 24 GB+ steps up to larger chat models with richer writing. How much the writing actually changes tier to tier — from our own blind tests — is in writing quality by GPU tier. Full hardware ladder: system requirements.
Does "local" mean worse images?
No — local is the point. Fervyn's image engine is tuned per GPU tier, and the image-first design (character consistency, true multi-character scenes, a prose-to-layout extractor) is what sets it apart, not raw cloud horsepower.
Can I make it faster, or higher quality?
Yes — above the 8 GB floor, Settings lets you pick a Speed preset (the chat model and the image model stay loaded together, so there's no wait between a reply and its picture) or a Quality preset (a larger chat model and a higher-fidelity image engine, swapped in as needed — a few seconds more per picture). On a 32 GB card you can run both at full quality at once — no swap pause. Details on system requirements.
Can I import my existing characters?
Yes — Fervyn reads standard TavernCard / Chub character cards.
Is the source open?
The companion repository is open so the privacy guarantee is auditable. (Image models and some assets are distributed separately.)
Is it available everywhere?
Not yet everywhere. Some regions are excluded at download while legal review completes. We don't publish a compliance claim for any specific jurisdiction. 18+ only.