AI image generator with multi-reference fusion and advanced character consistency
Translate manga. Keep the art.
Stage
Not given
Not given
Founded
Not given
Not given
Based in
Not given
Not given
Pricing model
Freemium
Paid
Pricing
Free credits available upon signup. Subscription options include monthly subscriptions with automatic monthly billing and annual subscriptions with up to 50% savings. One-time credit packs available for purchase. Specific plan names and prices not stated on the pages.
Starter: $4.99 for 100 pages ($0.05 per page); Pro: $19.90 for 1,000 pages ($0.02 per page); Max: $99.90 for 10,000 pages ($0.01 per page). New accounts include 3 free pages.
Key features
Text-to-image generation with flux text to image prompts
Image-to-image editing and refinement
Output up to 4MP with flexible aspect ratios
90% text legibility with multilingual typography support
Physics-aware lighting and realistic rendering
Three model variants: Pro, Flex, and Dev
Generation in under 10 seconds
JSON-structured prompts for precise control
FP8 quantization for 40% VRAM savings
Detects speech bubbles, captions and vertical Japanese text
Erases original lettering and repairs artwork
Typesets translation inside original bubbles
Three translation styles: Standard, Keep honorifics, Faithful
Glossary for character names and terms
Supports 27 target languages
Reads Japanese, Korean, Chinese and English automatically
Batch processing up to 20 pages at once
Supports JPG, PNG and WebP formats up to 10 MB
What makes it different
Eliminates stochastic drift in edits that plague other tools
32-billion parameter rectified flow matching architecture for physics-aware generation
Professional-grade output up to 4MP for print-ready results
90% typography accuracy for multilingual layouts and text rendering
One page translation uses one credit regardless of size or text amount
Credits never expire and failed pages are refunded automatically
No subscription required; one-time purchases
Built specifically for comic pages, not paragraph text
Finished image output with text in place, not text list