polym-eval-generate-veo

Name: polym-eval-generate-veo
Availability: InStock
Author: byteplus-sa

Generate short Veo videos from prompts and first-frame images via Vertex AI or AIDP.

8|Updated May 13, 2026

Search Tags:#video generation #text-to-video #image-to-video #vertex ai #veo #aidp #mp4 output

Ready to run

npx skills add https://github.com/byteplus-sa/polym --skill polym-eval-generate-veo

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: polym-eval-generate-veo
Source: https://github.com/byteplus-sa/polym/tree/main/skills/polym-eval-generate-veo
Command: npx skills add https://github.com/byteplus-sa/polym --skill polym-eval-generate-veo

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

It eliminates the manual effort of setting up video generation pipelines by letting you produce Veo (text-to-video and image-to-video) outputs from simple inputs like a prompt and optional first-frame reference.

Core Features & Use Cases

Prompt-based video generation: Create short videos from text prompts using Google Veo via Vertex AI or ByteDance AIDP.
First-frame / image-to-video workflows: Turn a reference image into a video by auto-detecting the aspect ratio when needed.
Operational configuration & outputs: Controls for duration, aspect ratio, model selection, and outputs to a local mp4 path with an optional network URL flow via TOS.

Quick Start

Run the Vertex AI text-to-video path by executing python scripts/generate.py --prompt "A golden retriever running through a meadow at sunset".

Frequently Asked Questions about polym-eval-generate-veo

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

FAQPage Schema

How do I generate text-to-video and image-to-video outputs using Google Veo?▼

You can generate Veo videos by running a Python script with your text prompt and an optional first-frame reference image. The Skill processes these inputs via Vertex AI or AIDP backends to return a short mp4 artifact.

What environment configuration is required to run Veo video generation pipelines?▼

Veo video generation requires specific environment variables depending on your chosen backend. You must configure VERTEX_PROJECT and VERTEX_LOCATION for Vertex AI, or AIDP_AK for the AIDP backend.

Can I use a reference image to generate a video and auto-detect the aspect ratio?▼

Yes, you can supply a reference image for image-to-video workflows. The Skill automatically detects the required aspect ratio from the first-frame image to generate the matching mp4 output.

Does Veo video generation support different durations and aspect ratios for mp4 outputs?▼

Yes, the Veo generation process provides controls for duration and aspect ratio. This ensures predictable mp4 outputs tailored to your specific text-to-video or image-to-video workflow requirements.

How do I get a network URL for my generated Veo mp4 video file?▼

To obtain a network URL for your generated Veo mp4 video, you must configure TOS alongside your backend environment variables. The Skill returns this network URL when TOS is properly set up.