# MOSS-Image

Text to image with legible text in English and Chinese and JSON layouts for placing text and objects, 1024x1024 by default.

Model id: `moss-image`
Quantization: bf16
Regions: ATL
Hugging Face: [OpenMOSS-Team/MOSS-Image-SFT](https://huggingface.co/OpenMOSS-Team/MOSS-Image-SFT)

## Pricing

- Image: $0.02 / megapixel

## Inputs

- text

## Output: image

| Parameter | Type | Values | Default |
| --- | --- | --- | --- |
| `size` | any |  |  |
| `aspect_ratio` | string | 1:1, 4:3, 3:4, 3:2, 2:3, 16:9, 9:16, 21:9, 9:21 | 1:1 |
| `resolution` | string | 768, 1K, 1.5K, 2K |  |
| `n` | integer | 1–4 | 1 |
| `response_format` | string | url, b64_json | url |
| `seed` | integer |  |  |
| `steps` | integer | 20–50 | 50 |
| `output_format` | string | png, jpeg, webp | png |
| `output_compression` | integer | 0–100 |  |
| `background` | string | auto, opaque | auto |

## Endpoints

- [`POST /v1/images/generations`](/docs/reference/create-image.md): Create Image
- [`POST /v1/images`](/docs/reference/create-image-openrouter.md): Generate Image (OpenRouter)
- [`POST /v1/images/edits`](/docs/reference/create-image-edit.md): Edit Image