Skip to main content

Hunyuan Image Generation Model Upgrade Guide

The image generation credits granted by the Mini Program Growth Initiative Phase 1 currently use the hunyuan-image model. This model will be shut down at 00:00 on June 30, 2026.

Please complete the migration before the shutdown date by switching to hunyuan-image-v3.0-v1.0.4. If the migration is not completed before the deadline, the system will automatically switch to hunyuan-image-v3.0-v1.0.4.

After the upgrade, the image generation credits from the Mini Program Growth Initiative can continue to be used with hunyuan-image-v3.0-v1.0.4.

Affected Models​

Old Model NameNew Model Name
hunyuan-imagehunyuan-image-v3.0-v1.0.4

The new model hunyuan-image-v3.0-v1.0.4 supports custom aspect ratios, prompt rewriting, and a thinking-based rewriting mode, resulting in higher quality image generation.

Migration Guide​

Image generation is only supported in Cloud Functions (Node.js SDK). Updating the model parameter in generateImage is all that is needed to complete the upgrade:

Using @cloudbase/node-sdk​

Before:

const cloudbase = require('@cloudbase/node-sdk');
const app = cloudbase.init({ env: process.env.ENV_ID });
const ai = app.ai();

exports.main = async (event) => {
  const { prompt } = event;

  const imageModel = ai.createImageModel('hunyuan-image');
  const res = await imageModel.generateImage({
    model: 'hunyuan-image',
    prompt,
  });

  return { url: res.data[0].url };
};

After:

const cloudbase = require('@cloudbase/node-sdk');
const app = cloudbase.init({ env: process.env.ENV_ID });
const ai = app.ai();

exports.main = async (event) => {
  const { prompt } = event;

  const imageModel = ai.createImageModel('hunyuan-image');
  const res = await imageModel.generateImage({
    model: 'hunyuan-image-v3.0-v1.0.4',
    prompt,
    size: '1024x1024',                    // optional, supports custom aspect ratio
    revise: { value: true },              // optional, enable prompt rewriting
    enable_thinking: { value: true },     // optional, enable thinking mode
  });

  return { url: res.data[0].url };
};

Using wx-server-sdk​

Before:

const cloud = require('wx-server-sdk')
cloud.init({ env: cloud.DYNAMIC_CURRENT_ENV })

exports.main = async (event, context) => {
  const imageModel = cloud.ai().createImageModel('hunyuan-image')
  const res = await imageModel.generateImage({
    model: 'hunyuan-image',
    prompt: event.prompt,
  })
  return { url: res.data[0].url }
}

After:

const cloud = require('wx-server-sdk')

cloud.init({
  env: cloud.DYNAMIC_CURRENT_ENV,
  timeout: 150000,   // HTTP request timeout 150s, overrides the ~15s default
})

const ALLOWED_SIZES = ['1024x1024', '1280x720', '720x1280', '1280x1280']

exports.main = async (event, context) => {
  const prompt = (event.prompt || '').trim()
  if (!prompt) {
    return { error: 'Please enter a prompt' }
  }
  if (prompt.length > 500) {
    return { error: 'Prompt must be 500 characters or less' }
  }

  const size = ALLOWED_SIZES.includes(event.size) ? event.size : '1024x1024'

  const imageModel = cloud.ai().createImageModel('hunyuan-image')
  const res = await imageModel.generateImage({
    model: 'hunyuan-image-v3.0-v1.0.4',
    prompt,
    size,
    revise: { value: false },
    enable_thinking: { value: false },
  })

  return {
    url: res.data[0].url,                         // Image URL, valid for 24 hours
    revisedPrompt: res.data[0].revised_prompt,    // Optimized prompt
  }
}

The new model supports additional parameters. See Node SDK and wx-server-sdk for details.

Getting Help​

If you run into issues during migration, visit the CloudBase Community or submit a support ticket for 1-on-1 assistance.