polym-eval-generate-seedream

Generate images from Seedream text prompts via the BytePlus Ark images/generations API.

8|Updated May 13, 2026
One-click install
npx skills add https://github.com/byteplus-sa/polym --skill polym-eval-generate-seedream
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: polym-eval-generate-seedream
Source: https://github.com/byteplus-sa/polym/tree/main/skills/polym-eval-generate-seedream
Command: npx skills add https://github.com/byteplus-sa/polym --skill polym-eval-generate-seedream

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires requests, tqdm, python-dotenv, tos, and includes scripts (resource) components.

What problem does it solve?

This Skill helps you produce high-quality AI images from text prompts without manually wiring up BytePlus Ark API calls or handling generation parameters.

Core Features & Use Cases

  • Environment-validated image generation: Ensures required BytePlus credentials exist (and auto-creates/updates the .env file when missing) before any generation runs.
  • Seedream image creation with controls: Supports configurable model, output size, guidance scale, random seed, watermark, and response formats (URL or base64).
  • Reference image support and batch-style generation: Lets you include reference images (local or URL), with optional sequential multi-panel generation for consistent outputs.

Quick Start

Run the generator with a prompt by executing: python skills/polym-eval-generate-seedream/scripts/generate.py --prompt "A serene mountain landscape at sunset with golden clouds" --output generated/mountain.png.

Frequently Asked Questions about polym-eval-generate-seedream

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate images from text prompts using the BytePlus Ark API?

To generate images from text prompts using the BytePlus Ark API, configure your environment variables with BYTEPLUS_API_KEY and BYTEPLUS_BASE_URL, then run the generation script. The script validates credentials and outputs a saved image file along with a direct URL or base64 payload.

Can I use reference images for AI image generation with Seedream?

Yes, you can use reference images for AI image generation with Seedream by providing local file paths or URLs. This conditions the model on your provided visuals, and you can use sequential multi-panel generation to ensure consistent outputs across multiple images.

Do I need TOS credentials to create Seedream images?

You do not strictly need TOS credentials to create Seedream images; BYTEPLUS_API_KEY and BYTEPLUS_BASE_URL are the required environment variables. TOS credentials are optional and only needed if your specific workflow requires object storage integration for image handling.

How do I ensure reproducible AI image generation results?

To ensure reproducible AI image generation results, use the random seed parameter in the generation script. Setting a fixed seed allows the model to produce identical visual outputs from the same text prompt and generation parameters across multiple runs.

What parameters can I configure for BytePlus Ark image generation?

You can configure model selection, output sizing, guidance scale, random seed, watermark inclusion, and response formats (URL or base64) for BytePlus Ark image generation. These parameters are passed as command-line arguments to the generation script.

What's the best way to automate sequential multi-panel image generation for product visuals?

The best way to automate sequential multi-panel image generation for product visuals is using the script's sequential generation feature with reference images. This ensures visual consistency across panels by conditioning the model on the same reference inputs throughout the workflow.