Text-to-image on MOSS-Image, OpenMOSS's open 5-billion-parameter model that sets long, exact text into posters, covers and signs. Prompts up to 12,000 characters, an experimental JSON layout, 15 aspect ratios, three resolution tiers, and no content filter.
Workbench
These examples were written for this endpoint. Replace the prompt and sample media URLs with your own accessible inputs, then check the current quote and model availability before submitting.
Outputmeridian-film-poster Every string is quoted: the title, the tagline, the cast and director line and the release date. At 1.5k the small type at the bottom stayed readable.
- 21
- 1.5k
- 16:9
Outputvelvet-magazine-cover The masthead, three cover lines and the price line are each quoted and placed. 1:2 suits a tall cover.
- 7
- 1k
- 1:2
Outputmidnight-sessions-layout The prompt is a layout JSON with four boxes, calculated for 1168 x 880. Text and dancer land where the boxes put them. The format is experimental.
- 7
- 1k
- 4:3
Outputnoir-perfume-ad A quoted wordmark, a spaced-capitals line and an italic tagline at 2k. On another seed the letters of NOIR fused, so keep a spare seed.
- 7
- 2k
- 9:16
Outputmotel-paradiso-banner The sign text is quoted, and 21:9 gives a wide, filmic frame. One seed, no retries.
- 7
- 1k
- 21:9
This example starts at $0.024 per image. The Playground updates the estimate when you change the settings.