Image Generation

Generate images from text descriptions, with support for long text and text rendering, for poster design, illustration, and visual content creation.

1. Create an API Key

   Open the API Key Management Platform, create an API Key, and save it as an environment variable. Do not write the API Key into client-side code or commit it to a repository.

   export MOSS_API_KEY='YOUR_API_KEY'

2. Prepare the request input

   Provide the text description, the image size, and the model ID. Currently only one image is generated per request, and streaming output is not supported.

   FieldTypeRequirementDescription

   model: stringRequired

   Model ID. Use the stable model ID for regular calls, or a snapshot model ID when you need to reproduce results from a fixed version.

   Show 4 possible values

   "moss-image-base-1.0"

   Stable model ID.

   "moss-image-base-1.0-2026-10-09"

   Snapshot model ID for pinning the model version.

   "moss-image-sft-1.0"

   Stable model ID (SFT version).

   "moss-image-sft-1.0-2026-10-09"

   Snapshot model ID (SFT version) for pinning the model version.

   prompt: stringRequired

   The text description of the image to generate; a non-empty string of at most 4,000 characters.

   size: stringOptional

   Image size in the format widthxheight. Each side must be 256–2048 and a multiple of 16; the total pixel count must not exceed 2097152.

   n: integerOptional

   Number of images to generate; only 1 is supported.

   response_format: stringOptional

   Only url is supported; b64_json is not supported.

   {
     "model": "moss-image-sft-1.0",
     "prompt": "An orange cat sitting by the window, a garden after the rain outside, soft natural light, warm and delicate photographic style.",
     "negative_prompt": "nsfw, worst quality, bad anatomy, watermark, jpeg artifact",
     "size": "1024x1024",
     "n": 1,
     "response_format": "url"
   }

3. Send the image generation request

   Send the request to /v1/images/generations. When generation completes, the endpoint returns JSON.

   curl --max-time 900 -X POST "https://api.mosi.cn/v1/images/generations" \
     -H "Authorization: Bearer $MOSS_API_KEY" \
     -H "Content-Type: application/json" \
     -d '{
     "model": "moss-image-sft-1.0",
     "prompt": "An orange cat sitting by the window, a garden after the rain outside, soft natural light, warm and delicate photographic style.",
     "negative_prompt": "nsfw, worst quality, bad anatomy, watermark, jpeg artifact",
     "size": "1024x1024",
     "n": 1,
     "response_format": "url"
   }'

   import os
   import requests
   
   resp = requests.post(
       "https://api.mosi.cn/v1/images/generations",
       headers={
           "Authorization": f"Bearer {os.environ['MOSS_API_KEY']}",
           "Content-Type": "application/json",
       },
       timeout=900,
       json={
           "model": "moss-image-sft-1.0",
           "prompt": "An orange cat sitting by the window, a garden after the rain outside, soft natural light, warm and delicate photographic style.",
           "negative_prompt": "nsfw, worst quality, bad anatomy, watermark, jpeg artifact",
           "size": "1024x1024",
           "n": 1,
           "response_format": "url"
       },
   )
   
   print(resp.status_code)
   print(resp.json())

   const response = await fetch("https://api.mosi.cn/v1/images/generations", {
     method: "POST",
     signal: AbortSignal.timeout(900_000),
     headers: {
       Authorization: `Bearer ${process.env.MOSS_API_KEY}`,
       "Content-Type": "application/json",
     },
     body: JSON.stringify({
       "model": "moss-image-sft-1.0",
       "prompt": "An orange cat sitting by the window, a garden after the rain outside, soft natural light, warm and delicate photographic style.",
       "negative_prompt": "nsfw, worst quality, bad anatomy, watermark, jpeg artifact",
       "size": "1024x1024",
       "n": 1,
       "response_format": "url"
     }),
   });
   
   console.log(response.status);
   console.log(await response.json());

   The example timeout is 900 seconds; it does not represent a fixed server-side timeout.

4. Get the image result

   On success, read the image address from data[0].url in the response; you can open or download the PNG directly. usage.generated_images, usage.output_width, and usage.output_height report the number of images actually generated, the image width, and the image height.

   # Replace the example URL with data[0].url
   curl --fail --location "https://example.com/generated-image.png" \
     --output image.png

   import requests
   
   # Replace the example URL with data[0].url
   image_url = "https://example.com/generated-image.png"
   resp = requests.get(image_url)
   resp.raise_for_status()
   
   with open("image.png", "wb") as file:
       file.write(resp.content)
   
   print("Image saved to image.png")

   import { writeFile } from "node:fs/promises";
   
   // Replace the example URL with data[0].url
   const imageUrl = "https://example.com/generated-image.png";
   const response = await fetch(imageUrl);
   if (!response.ok) {
     throw new Error(`Download failed: ${response.status}`);
   }
   
   await writeFile("image.png", Buffer.from(await response.arrayBuffer()));
   console.log("Image saved to image.png");

5. Verify success

   Success means you receive the result image URL and can view or download the generated PNG image.

Next steps