MOSS Image BaseBeta

MOSS Image Base: AI Text to Image Generator

Generate images from text with MOSS Image Base. Explore compositions, illustrations and bilingual visual ideas in square, portrait or landscape formats.

4 credits / imageYour credits work across all models.

Your output

A little input. A new possibility.

Your finished audio, image or text will appear here.

01 / Inside the model

Make room for a visual first idea.

MOSS Image Base turns a written prompt into an image. The official MOSS-Image family uses a 5B backbone and explores Chinese and English text rendering, structured layouts and a range of visual styles. In this studio, describe your scene and choose a square, portrait or landscape canvas. Base is the lower-credit image option, useful for exploring compositions before deciding what to refine.

Text-led composition

Describe the subject, setting, lighting and visual treatment.

Three canvas shapes

Choose square, portrait or landscape output.

Explore before refining

Generate a visual draft, then adjust the prompt for the next idea.

02 / Examples & listening room

See it. Hear it. Get a feel for it.

Explore the official model

Official examples are credited to OpenMOSS. Studio samples were generated by us with the named API model. These are saved previews, not a result of your current settings. Research capabilities and available studio controls may differ.

Base checkpoint · Chinese text rendering
Official OpenMOSS example

Geometric visual composition
Official OpenMOSS example

03 / Use cases

Where it fits.

Moodboard directions

Explore an aesthetic for a campaign or creative brief.

Visual concepts

Try compositions for an illustration, product scene or poster draft.

04 / Questions

A few things worth knowing.

What does Image Base cost?

4 credits per generated image. The selected preset canvas shape does not change that rate.

How does Base differ from SFT?

They are separate checkpoints in the official MOSS-Image family. Base costs 4 credits here; SFT costs 6. Compare the same prompt to choose your preferred output.

Can the model draw readable text?

The family is designed for Chinese and English text rendering, but spelling and layout can still be wrong. Inspect the actual image.

Made for your next idea

Choose a model, add your input and review the credit cost. Successful results stay in your account history; AI output may be inaccurate.

Text is rounded up to a whole credit. Audio is billed per started minute. Live voice uses one credit per 6 seconds.

Failed tasks are refunded automatically.

Result links can expire. Download files you want to keep.

Model documentation