MOSS-Image SFT · Apache-2.0

MossImage: Free Online AI Image Generator with Exact Text Placement

Type a prompt, place the exact words where you want them, and get a print-ready image in seconds.

Start creating — free Browse examples

No credit card · English & Chinese text

MOSS-Image exhibition poster for The Moss Archive with precise English and Chinese typography — Bilingual exhibition poster · long-form type

MOSS-Image night market event poster with bold bilingual display type — Event poster · display type and metadata

MOSS-Image product advertisement with a ceramic mug and a $24 price badge — Product ad · headline, price and product shot

Create an image

PreviewMOSS-Image SFT

Sample MOSS-Image output: a bilingual Moss & Ember coffee packaging label

Sample output · Packaging

Your image will appear here

Pick a tier, set a ratio, and press Generate. Drafts render square at 512 px.

Runs MOSS-Image SFT. Weights are Apache-2.0.

Examples

What MossImage can make

Eight recent MossImage renders, one per prompt in the showcase. Every prompt is copyable and ready to try.

Cinematic key art poster for the fictional film The Last Signal, with a radio astronomer and a dense billing block — The Last Signal — film key art

COMMON magazine cover with a model in a knit sweater, cover lines and a barcode — COMMON — magazine cover

Still Mountain osmanthus oolong tea pouch with Chinese and English label text — Still Mountain — bilingual tea pack

Northline GT billboard on a rainy Chicago street with headline and legal copy — Northline — rainy billboard

The Blue Counter diner menu with two columns of dishes, prices and a legal line — The Blue Counter — diner menu

Pulse SaaS revenue dashboard with a chart, summary cards and a top accounts table — Pulse — SaaS dashboard UI

Northbound music festival poster with an orange risograph sunset and 27 artist names — Northbound — festival lineup

Night Shift comic page with six panels, speech balloons and orange risograph ink — Night Shift — comic page

Images generated with MossImage (MOSS-Image-SFT).

Browse the prompt library

Workflow

Three steps to a readable image

From a plain prompt to a print-ready file in three steps. No install and no design software.

Describe the scene and quote the words

Write the setting, then paste the exact headline, menu or caption. Quotation marks and line breaks survive the trip to the model.

Pick a render tier and ratio

Draft checks spelling at 512 px. Final renders at about 1 megapixel. HD doubles that to about 2 megapixels, and Max goes up to 4 megapixels.

Render, download, iterate

Download the PNG. If a line breaks badly, tighten the wording in the prompt and render again.

Try it in the generator →See the full walkthrough

Built for text

Why MossImage

A text-first image model, with the controls a real design workflow needs.

Text that actually renders

Long English and Chinese strings stay legible — headlines, menus and paragraphs, not gibberish.

See the model comparison

Layout you control

Pin text and objects to exact positions with a JSON layout, or start from a template.

How layout control works

Run it locally

Apache-2.0 weights and a ComfyUI workflow if you want the model on your own GPU.

Download the workflow

Pricing

Simple plans for every workflow

Monthly plans, annual plans that save 20%, and one-time credit packs that never expire. Every paid image is watermark-free.

All prices in USD · charged by Waffo

Subscription

Lite

50 credits / month · ~16 Final renders

$9.99/ month

50 credits·$0.20 / credit·/ month

Watermark-free exports
Commercial use license
Credits refresh each billing cycle

Secure checkout · cancel anytime

Pro

120 credits / month · ~40 Final renders

$19.99/ month

120 credits·$0.17 / credit·/ month

Watermark-free exports
Commercial use license
Credits refresh each billing cycle
Priority generation queue

Secure checkout · cancel anytime

BEST VALUE

Subscription

Studio

300 credits / month · ~100 Final renders

$39.99/ month

300 credits·$0.13 / credit·/ month

Watermark-free exports
Commercial use license
Credits refresh each billing cycle
Priority generation queue

Secure checkout · cancel anytime

See full pricing, credit costs and FAQ →

FAQ

Questions, answered

What is MossImage?

MossImage is a hosted, browser-based image generator built on MOSS-Image, the open-source model released by OpenMOSS on 16 October 2026 under Apache-2.0. It is designed for images that carry real text: posters, menus, signs, packaging and UI mockups.

Is MossImage free?

Yes. Signed-in accounts get two free credits per day when they open the site. A Draft costs 1 credit, so that is two Draft renders. Free-account renders carry a MossImage watermark; any subscription or one-time credit pack removes it. No credit card is needed to try the free tier.

Do I need a subscription?

No. A subscription is optional. You can use the free daily credits, or buy a one-time credit pack that never expires. Subscriptions are for people who want a recurring credit refresh and watermark-free exports every month; one-time packs also remove the watermark.

How much does MossImage cost?

Monthly plans start at $9.99 for 50 credits, yearly plans start at $95.88 for 600 credits, and one-time packs start at $4.99 for 15 credits. The free tier gives 2 credits per day. All prices are in USD and billed by Waffo.

Do I need to install anything?

No. Type a prompt, choose a tier and ratio, and generate in the browser. If you want to run the model on your own machine, use the ComfyUI workflow or the official Apache-2.0 weights.

Read all questions →