Deploy Stable Diffusion on Alibaba Cloud Function Compute for Light‑Shadow Text Effects
This guide walks you through registering an Alibaba Cloud account, enabling Function Compute and NAS, deploying a Stable Diffusion instance, adding ControlNet, and handling common cost questions, enabling you to create light‑shadow text images in just a few minutes.
Introduction
Stable Diffusion has become popular for generating eye‑catching visual effects, and creating light‑shadow text is a useful advanced technique. This tutorial explains how to deploy Stable Diffusion on Alibaba Cloud Function Compute and use ControlNet to produce such effects.
Resource Preparation
Register and log in to an Alibaba Cloud account.
Enable Function Compute and claim the free trial quota.
Enable the NAS (Network Attached Storage) service and claim its free trial quota.
Deploying Stable Diffusion on the Cloud
Select the "AI Digital Painting – Stable Diffusion Custom Template" and click Create Immediately at https://fcnext.console.aliyun.com/applications/create.
Perform basic configuration: confirm that Function Compute (FC) and NAS are enabled and accept the default role name.
Advanced configuration: choose a region, select a drawing type, and grant the RAM role APN permission for the first deployment.
Create and deploy the default environment, acknowledging that any usage beyond the free quota will be billed on a pay‑as‑you‑go basis.
After deployment, two URLs are provided: one for the Stable Diffusion WebUI (prefixed with sd) and another for the admin backend (prefixed with admin) where models are uploaded.
Preparing the Large Model
Download the majicMix realistic model from the model library; it will be stored in your NAS and may incur storage fees. After downloading, restart Stable Diffusion to load the model.
Preparing ControlNet
Download the ControlNet plugin from its GitHub repository: https://github.com/Mikubill/sd-webui-controlnet.
Upload the downloaded ZIP file to the admin backend.
Refresh the directory, unzip the plugin, and move its contents to the parent directory.
Delete the original ZIP and temporary folders.
Reload the Stable Diffusion WebUI; the ControlNet configuration panel should appear.
Creating Light‑Shadow Text
Prepare a high‑contrast black‑background white‑text image using PowerPoint or Photoshop.
Set prompt words (e.g., city, building, neon light ) and negative prompts (e.g., EasyNegative, worst quality ). Use around 20 sampling steps for clarity.
Upload the text image to ControlNet, enable All , set preprocessing to None , and select the brightness model.
Adjust parameters: Control Weight ≈ 0.4, Starting Control Step 0–0.2, Ending Control Step 0.5–0.7.
Common Cost Issues
Q1: Why is there a small Function Compute charge despite the free trial? The free trial does not cover outbound internet traffic generated by Stable Diffusion. For example, generating 100 images at 5 MB each consumes 500 MB, costing about ¥0.25 (¥0.50/GB).
Q2: Why does NAS storage incur fees after using the free quota? The free NAS quota applies to either 50 GB of capacity‑type storage or 9.15 GB of performance‑type storage. This tutorial uses performance‑type NAS, so only 9.15 GB is free; any excess is billed at ¥1.8 per GB per month.
References
Alibaba Cloud account login: https://account.aliyun.com/login/login.htm Function Compute free trial:
https://free.aliyun.com/?pipCode=fc&spm=5176.137990.J_5253785160.5.36551608plfTh6NAS free trial: https://free.aliyun.com/?product=9657388&crowd=personal Function Compute billing overview:
https://help.aliyun.com/zh/fc/product-overview/billing-overviewSigned-in readers can open the original source through BestHub's protected redirect.
This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactand we will review it promptly.
Alibaba Cloud Native
We publish cloud-native tech news, curate in-depth content, host regular events and live streams, and share Alibaba product and user case studies. Join us to explore and share the cloud-native insights you need.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.
