Tencent Cloud Hunyuan Text to Image

What problem it solves

When an agent needs to turn a user's text description into an AI-generated image, calling a cloud image API directly involves job submission, status polling, and result URL extraction. TencentCloud Aiart TextToImage wraps Tencent Cloud's Hunyuan text-to-image service into executable scripts, making it a tool for image generation rather than just describing the image.

How it works

• Default entrypoint: use main.py to submit the job and poll for the result automatically. Prefer this mode when the user does not specify a workflow.
• Step-by-step control: submit_job.py only submits the job and returns a JobId; query_job.py queries or polls by JobId, which fits asynchronous pipelines.
• Key parameters: --images accepts up to 3 reference image URLs in jpg/jpeg/png/webp, with each base64 payload under 10MB; --resolution sets width and height between 512 and 2048 pixels, with total area no more than 1024×1024; --seed fixes reproducibility; --revise enables prompt revision by default and adds about 20 seconds.
• Runtime setup: it requires Tencent Cloud API credentials and can install Python dependencies automatically; keep credentials in environment variables rather than hard-coding them.

Caveats

Generated image URLs expire after 1 hour, so save them promptly. Prompt revision increases latency. If the call fails, return a clear error instead of fabricating an image URL or describing the result.