polym-eval-generate-gemini

Generate PNG images from text prompts using Google Gemini/Imagen or ByteDance AIDP backends.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Generating images from prompts (and optionally editing/steering with reference images) without manually wiring up Vertex AI, AI Studio, or an internal image-generation API.

Core Features & Use Cases

  • Prompt-driven image generation: Create new images from a text prompt using Google Gemini/Imagen backends.
  • Optional reference-image workflows: Provide local or URL reference images to guide style/composition for generation/editing.
  • Backend choice with guidance: Use google (Vertex AI + AI Studio with fallback) or aidp (ByteDance AIDP multimodal/crawl) depending on availability and speed needs.

Quick Start

Use the skill to generate an image by running: python skills/polym-eval-generate-gemini/scripts/generate.py --prompt "A serene mountain landscape at sunset".

Frequently Asked Questions about polym-eval-generate-gemini

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

FAQPage Schema
How do I generate images from text prompts using Google Gemini or Vertex AI?

To generate images from text prompts using Gemini or Vertex AI, you run a Python script that calls Google Gemini or Imagen backends. Simply provide a text prompt to automatically create new images without manually wiring up the APIs.

Can I use reference images to guide image generation and editing with Imagen?

You can use reference images to guide image generation and editing with Imagen by providing local files or URLs. This steers the style and composition of the generated output to match your provided visual references.

Do I need an API key to generate images with Google AI Studio or ByteDance AIDP?

You need an API key to generate images with Google AI Studio or ByteDance AIDP. A configured .env file containing your GOOGLE_AI_STUDIO_API_KEY and/or AIDP_API_KEY is required for backend authentication.

What is the difference between using Google Vertex AI and ByteDance AIDP for image generation?

The difference between using Google Vertex AI and ByteDance AIDP for image generation is backend availability and speed. Google uses Vertex AI with AI Studio fallback, while AIDP offers a multimodal crawl backend for different performance needs.

How do I output generated images as PNG files with a presigned URL via TOS?

To output generated images as PNG files with a presigned URL via TOS, the generation script automatically saves the results locally and optionally generates a presigned URL. This allows immediate access and sharing of the created PNG files.