Hosted Server

The managed backend selected by default. Click Translate and it just works — no API keys, no setup.

Overview

This is the default backend for Manga Translator Z. When you install the extension, it's already configured to talk to our hosted FastAPI service at mtz-api.peakproductivity.online. The server runs the complete translation pipeline — text detection, OCR, GPT translation, inpainting (painting over the original text), and text rendering — and returns a fully translated image that the extension swaps in place of the original.

Pricing

The free tier includes a weekly page quota. When you hit it, three upgrade tiers are available from the popup's Account tab:

Tier	Price	Good for
Free	$0 / week	Casual readers — a few chapters per week
Day Pass	$0.99 / 24h	Binge-reading a whole series in one sitting
Starter	$4.99 / month	Regular readers following ongoing titles
Power	$9.99 / month	Heavy readers and scanlation reference

Billing is handled through Stripe. Authentication uses magic-link email — no passwords to manage. Usage counts only against successful translations — failed pages never burn quota.

What the server does

1. Text Detection — CRAFT / PaddleOCR finds bubble and SFX regions
2. OCR — Recognizes source-language text in each region
3. Translation — Translates via GPT with glossary + tone context
4. Inpainting — Paints over the original text pixels
5. Text Rendering — Draws the translated text back onto the cleaned image

Progress streaming

Each translation streams progress back to the extension via Server-Sent Events (SSE). The progress overlay surfaces the current stage — detecting, OCR, translating, inpainting, rendering, complete — so you always know what's happening.

Self-hosting (advanced)

The server is open source: the same FastAPI app that powers the hosted service can be run on your own machine or GPU box. Point the extension at your own URL via the Backend tab to use it.

1. Install dependencies:

   pip install -r requirements.txt

2. Configure environment: Copy .env.example to .env and fill in values
3. Start the server:

   python run.py

4. In the popup's Backend tab, select "FastAPI Server"
5. Replace the default URL with your server's URL and paste your API key
6. Click "Check Connection"