LLM
Ming-Image-0.1-Design
Ming-Image-0.1-DesignA 6B text-to-image model for UIs, infographics, posters, and other text-rich designs, with RGBA output and transparent backgrounds.
Model details
View repositoryinclusionAI’s Ming-Image-0.1-Design is a 6B-parameter text-to-image model built for UI mockups, infographics, posters, and other text-rich visual designs. It generates complete compositions at up to 2048 × 2048 resolution and supports RGBA output with transparent backgrounds, making it well suited for production-ready design assets.
Input
1import base64
2import os
3
4import requests
5
6model_id = "<your-model-id>"
7url = (
8 f"https://model-{model_id}.api.baseten.co"
9 "/environments/production/predict"
10)
11
12response = requests.post(
13 url,
14 headers={
15 "Authorization": f"Bearer {os.environ['BASETEN_API_KEY']}",
16 "Content-Type": "application/json",
17 },
18 json={
19 "prompt": (
20 "A polished product-launch poster for a developer platform, "
21 "bold typography, dark background, electric blue accents"
22 ),
23 "resolution": 2048,
24 "num_inference_steps": 12,
25 "guidance_scale": 1.0,
26 },
27 timeout=300,
28)
29response.raise_for_status()
30
31result = response.json()
32image_base64 = result["data"]
33
34with open("ming-image.png", "wb") as image_file:
35 image_file.write(base64.b64decode(image_base64))
36
37print("Saved image to ming-image.png")
