Your words.
Your composition.
Create text-rich images, posters, and graphics. Edit the wording in your prompt and guide where headlines and objects appear—right from your browser.
Edit the words or move an element, then generate a new image. This plans a new composition; it does not edit an uploaded photo.
Specify the words you want, the lettering style, and the surrounding scene. MOSS Image can render text as part of the image, from poster titles and product labels to longer passages.
Official OpenMOSS Base sample. Online generation uses SFT; results and settings differ.
Structured prompts describe the scene, visible words, objects, and bounding boxes. Change the wording or placement to request a new composition—useful for posters, panels, and information graphics.
Edit a text & layout prompt Official OpenMOSS layout example. Layout guidance is experimental, not pixel-perfect design editing.Examples by OpenMOSS. View the official project ↗ · Always check spelling, names, prices, and element placement.
A headline, slogan, bottle, fruit, and price badge—two different compositions. Compare how their positions change. Colored boxes show the requested regions; the reference pair uses the same seed.
Full sizeHeadline above; bottle left; price badge and lemons right.
{
"summary": {
"meta": [
"poster"
],
"canvas": "single",
"caption": "A sunny yellow lemon-soda advertisement with navy lettering, a cold glass bottle, lemons, and a circular price badge. Crisp commercial photography and strong typography."
},
"items": [
{
"label": "text",
"caption": "Large navy headline.",
"bbox": [
-448,
-448,
-320,
448
],
"text": "SUNNY SODA"
},
{
"label": "text",
"caption": "Smaller navy slogan.",
"bbox": [
-304,
-352,
-240,
352
],
"text": "Squeezed today. Gone by noon."
},
{
"label": "object",
"caption": "A clear lemon-soda bottle with condensation and a yellow cap.",
"bbox": [
-192,
-416,
448,
-64
]
},
{
"label": "object",
"caption": "Fresh lemons, including one sliced lemon, with green leaves.",
"bbox": [
192,
32,
448,
448
]
},
{
"label": "text",
"caption": "Navy text inside a round white badge.",
"bbox": [
-160,
160,
64,
384
],
"text": "ONLY\n$3"
}
],
"bbox_coord_space": "abs_centralized",
"bbox_format": "yxyx"
}Adapted starter prompt, not the original generation recipe. Boxes are scaled to this studio’s supported image size. Outputs will vary.
Full sizeBottle right; badge upper left; headline and slogan below.
{
"summary": {
"meta": [
"poster"
],
"canvas": "single",
"caption": "A sunny yellow lemon-soda advertisement with navy lettering, a cold glass bottle, lemons, and a circular price badge. Crisp commercial photography and strong typography."
},
"items": [
{
"label": "text",
"caption": "Large navy headline.",
"bbox": [
320,
-448,
448,
448
],
"text": "SUNNY SODA"
},
{
"label": "text",
"caption": "Smaller navy slogan.",
"bbox": [
240,
-352,
304,
352
],
"text": "Squeezed today. Gone by noon."
},
{
"label": "object",
"caption": "A clear lemon-soda bottle with condensation and a yellow cap.",
"bbox": [
-448,
64,
192,
416
]
},
{
"label": "object",
"caption": "Fresh lemons, including one sliced lemon, with green leaves.",
"bbox": [
-64,
-448,
192,
-32
]
},
{
"label": "text",
"caption": "Navy text inside a round white badge.",
"bbox": [
-448,
-416,
-224,
-192
],
"text": "ONLY\n$3"
}
],
"bbox_coord_space": "abs_centralized",
"bbox_format": "yxyx"
}Adapted starter prompt, not the original generation recipe. Boxes are scaled to this studio’s supported image size. Outputs will vary.
Use structured layouts to organize dialogue, illustrations, headings, and smaller text blocks.
Full sizeSeparate panel and dialogue regions help organize a short visual story.
{
"summary": {
"meta": [
"comic"
],
"canvas": "multi_panel",
"caption": "A colorful three-panel garden comic. A friendly white robot with blue eyes finds a seed, waters a small potted plant, and celebrates a sunflower. Consistent character design and clear gutters."
},
"items": [
{
"label": "panel",
"caption": "The robot discovers a seed in a garden.",
"bbox": [
-429,
-656,
429,
-240
]
},
{
"label": "text",
"caption": "A clear outlined speech balloon.",
"bbox": [
-392,
-608,
-224,
-288
],
"text": "WHAT IS THIS?"
},
{
"label": "panel",
"caption": "The same robot waters a sprout in a terracotta pot.",
"bbox": [
-429,
-208,
429,
208
]
},
{
"label": "text",
"caption": "A clear outlined speech balloon.",
"bbox": [
-392,
-176,
-224,
176
],
"text": "GROW, LITTLE ONE."
},
{
"label": "panel",
"caption": "The robot cheers beside a tall sunflower.",
"bbox": [
-429,
240,
429,
656
]
},
{
"label": "text",
"caption": "A clear outlined speech balloon.",
"bbox": [
-392,
288,
-224,
608
],
"text": "WE DID IT!"
}
],
"bbox_coord_space": "abs_centralized",
"bbox_format": "yxyx"
}Adapted starter prompt, not the original generation recipe. Boxes are scaled to this studio’s supported image size. Outputs will vary.
Full sizeA shared scene, consistent characters, and English dialogue across four panels.
{
"summary": {
"meta": [
"comic"
],
"canvas": "multi_panel",
"caption": "A four-panel anime comic page in clean line art with soft cel shading, read left to right and top to bottom, with white gutters. An adult astronaut woman with short silver hair and a small round white helper robot share a ramen night aboard a space station."
},
"items": [
{
"label": "character",
"caption": "Top-left panel: the adult astronaut with short silver hair floats into the station galley holding two cup-noodle bowls.",
"bbox": [
-492,
-492,
-10,
-10
]
},
{
"label": "text",
"caption": "White comic speech bubble with bold black lettering.",
"text": "RAMEN NIGHT!",
"bbox": [
-472,
-461,
-378,
-204
]
},
{
"label": "character",
"caption": "Top-right panel: the small round white robot pours hot water into the bowls as steam drifts in zero gravity.",
"bbox": [
-492,
10,
-10,
492
]
},
{
"label": "text",
"caption": "White comic speech bubble with bold black lettering.",
"text": "3 MINUTES...",
"bbox": [
-472,
51,
-378,
308
]
},
{
"label": "character",
"caption": "Bottom-left panel: noodles float out of the bowls in zero gravity while the astronaut and the robot look surprised.",
"bbox": [
10,
-492,
492,
-10
]
},
{
"label": "text",
"caption": "White comic speech bubble with bold black lettering.",
"text": "UH-OH.",
"bbox": [
30,
-461,
123,
-266
]
},
{
"label": "character",
"caption": "Bottom-right panel: both laugh and catch the floating noodles with chopsticks, the glowing Earth in the window behind them.",
"bbox": [
10,
10,
492,
492
]
},
{
"label": "text",
"caption": "White comic speech bubble with bold black lettering.",
"text": "BEST NIGHT EVER.",
"bbox": [
30,
51,
123,
390
]
}
],
"bbox_coord_space": "abs_centralized",
"bbox_format": "yxyx"
}Published layout prompt from the source example, with its reference seed. The online model configuration differs; results will vary.
Full sizeArrange a title, central illustration, and four compact fact cards.
{
"summary": {
"meta": [
"poster"
],
"canvas": "single",
"caption": "A cream and honey-yellow educational poster about bees. Friendly bee illustration, dark legible lettering, warm accents, and four neatly aligned fact cards."
},
"items": [
{
"label": "text",
"caption": "Bold dark title.",
"bbox": [
-644,
-420,
-539,
420
],
"text": "THE LIFE OF A HONEYBEE"
},
{
"label": "text",
"caption": "Warm yellow subtitle.",
"bbox": [
-518,
-280,
-469,
280
],
"text": "Four facts from the hive"
},
{
"label": "object",
"caption": "A cheerful bee with translucent wings and a honeycomb accent.",
"bbox": [
-434,
-280,
98,
280
]
},
{
"label": "text",
"caption": "A white fact card with a dark heading and smaller detail.",
"bbox": [
140,
-420,
350,
-21
],
"text": "3 BODY PARTS\nHead, thorax and abdomen."
},
{
"label": "text",
"caption": "A white fact card with a dark heading and smaller detail.",
"bbox": [
140,
21,
350,
420
],
"text": "6 LEGS\nBack legs carry pollen."
},
{
"label": "text",
"caption": "A white fact card with a dark heading and smaller detail.",
"bbox": [
385,
-420,
595,
-21
],
"text": "5 EYES\nTwo compound eyes, three simple."
},
{
"label": "text",
"caption": "A white fact card with a dark heading and smaller detail.",
"bbox": [
385,
21,
595,
420
],
"text": "4 WINGS\nTwo pairs that hook together."
}
],
"bbox_coord_space": "abs_centralized",
"bbox_format": "yxyx"
}Adapted starter prompt, not the original generation recipe. Boxes are scaled to this studio’s supported image size. Outputs will vary.
Reference examples, not images generated in your session. Layout control is experimental; check text and placement before publishing. Editing a layout requests a new image—it does not edit an uploaded photo.
Explore real model outputs across text, layouts, illustration, and photography. Examples use different models and settings; results vary. Select a sample to prepare a prompt.
MOSS Image is an open-weight text-to-image model family from OpenMOSS. It combines image generation, text rendering, and experimental layout control in one model.
Use MOSS Studio to explore the hosted SFT model online. No local GPU or ComfyUI installation is needed to prepare a prompt.
Meet the OpenMOSS project| Model | What it is | In this studio |
|---|---|---|
| MOSS Image Base | Foundation text-to-image checkpoint; a starting point for further fine-tuning. | Model guide & official downloads |
| MOSS Image SFT | Fine-tuned from Base on curated data, sharing its architecture and inference pipeline. | Selected online model |
Base and SFT are two checkpoints in the MOSS Image family. Neither is presented here as a dedicated image-edit model.
Quote a headline, label, or line of dialogue. Describe the typography and style alongside the scene. Check the final lettering before publishing.
POSTERS · LABELS · SOCIAL GRAPHICSGuide where text and objects appear with structured layout prompts. Use the layout tab to plan your composition. Layout control is experimental, so results may vary.
HEADLINES · OBJECTS · COMPOSITIONExplore photographic scenes, illustrations, anime-inspired art, and graphic compositions. A detailed prompt gives the model more direction.
PHOTOGRAPHY · ILLUSTRATION · DESIGNUnderstand Base vs. SFT, and choose the right starting point for generation or fine-tuning.
Explore the model guide →LOCAL WORKFLOWFind the official plugin, model downloads, and workflow resources for local generation.
Open the ComfyUI guide →EDITING EXPLAINEDLearn what layout changes can do, and where reference-image editing needs a different workflow.
See image edit options →One-time credit packs, monthly plans, or annual memberships with about 25% savings.
Purchases are not open yet.
MOSS Image is an OpenMOSS text-to-image model family for creating images from natural-language prompts and structured layouts. MOSS Studio is an independent online experience, not the official OpenMOSS project.
The official model weights can be downloaded for local use under their published license; running them requires suitable hardware. New accounts receive 12 free credits, once per account. Additional online generation uses paid credits. You can explore examples and prepare prompts without paying. Generation availability is shown in the workspace.
Base is the foundation checkpoint and a starting point for further training. SFT is fine-tuned from Base on curated data and shares its architecture. This generator is configured for MOSS Image SFT; Base is not a second online model option here.
Describe the subject, style, lighting, and any words you want in the image, then choose an aspect ratio. Use Layout to guide headline placement. You can prepare everything without signing in; an account and sufficient credits will be needed to generate when the service opens.
MOSS Image supports text within images, including headlines, labels, and longer passages. Put the exact wording in quotes and describe the font style and placement. Lettering can still be imperfect, so check names, dates, spelling, and prices before publishing.
Native Layout Control uses a structured JSON prompt to describe objects, text, and bounding-box positions. Our Layout tab prepares guidance for headline placement. It guides composition rather than guaranteeing pixel-perfect placement.
Official Base and SFT downloads are available from OpenMOSS on Hugging Face. Our ComfyUI guide links to the official installation instructions and an example workflow. Download model files appropriate to your chosen loader; this website does not run a local ComfyUI workspace.
You do not need a local GPU for this online generator. Local hardware requirements depend on model precision, text encoder, resolution, offloading, and workflow. Use the official setup guide for your configuration rather than assuming a single VRAM number works for every version.
This generator accepts text and layout prompts, not uploaded photos or editing masks. Changing text or object positions in a layout generates a new image; it does not preserve an existing person, product, or background. Community editing workflows may have different capabilities.
This generator offers 1024 × 1024 square, 896 × 1344 portrait, and 1344 × 896 landscape images, one image per request. You can set a seed. Local MOSS workflows may offer additional resolutions and sampling settings.
The official SFT model card lists Apache-2.0 for its weights and bundled inference code. That does not clear third-party brands, characters, likenesses, or other rights in an output. Follow the model license and service terms, and review content rights before commercial use.
Square images use 10 credits; portrait and landscape use 12. Credit packs are one-time purchases, monthly plans add credits each paid month, and annual plans are billed once a year and issue the same monthly allowance for 12 months. The first grant arrives after verified payment; later grants follow the monthly billing anniversary. Canceling renewal does not remove grants for the already-paid year. Unused credits roll over. Confirmed generation failures restore reserved credits; an uncertain submission is not automatically resubmitted.
Explore first. Sign in when you’re ready to create.