← All models
MOSS-Image-SFT is a text-to-image model specializing in Chinese and English text rendering and structured layout control for precise text placement and object arrangement.
images
Details
- Accepts
- text
Pricing
- Price
- 6.25 cr / megapixel
Prices in credits (1 credit = $0.01).
Data schema
Input
| Field | Type | Description |
|---|---|---|
| seed | — | Random seed for reproducible generation. |
| promptrequired | string | Text prompt for image generation. Supports natural language, tag sequences, and JSON layouts. |
| sync_mode | boolean | Return images as data URIs instead of storing them in request history. |
| image_size | — | Output image size. Width and height are rounded to multiples of 16 between 512 and 2048 pixels. |
| num_images | integer | Number of images to generate. |
| output_format | stringenum: jpeg, png, webp | Output image format. |
| guidance_scale | number | Classifier-free guidance scale. |
| negative_prompt | string | Negative prompt. When empty, guidance uses transformer layer skipping for the negative branch. |
| num_inference_steps | integer | Number of denoising steps. |
| enable_safety_checker | boolean | Enable output content moderation. Disabling it requires account authorization; unauthorized requests are always checked, and images flagged as unsafe are returned as black images. |
| enable_prompt_expansion | boolean | Plan a structured layout from your prompt before generation. Preserves requested visible text and uses the selected image dimensions. If enhancement is unavailable, uses the original prompt. |
Output
| Field | Type | Description |
|---|---|---|
| seed | integer | Seed used for the first generated image. |
| images | array | Generated images. |
| actual_prompt | — | Expanded prompt used for generation, when enhancement succeeds. |
| has_nsfw_concepts | array | Whether each generated image contains unsafe content. |