Skybox AI Quick Start: Introduction To Building Application With Skybox AI

Overview

Skybox AI generates production-quality 360° equirectangular environments at 8K resolution (up to 16K) from text prompts in seconds. This guide helps application developers integrate Skybox AI via API, understand capacity planning, select appropriate plans, and avoid common implementation pitfalls.

API Documentation: api-documentation.blockadelabs.com

Quick Start Guides:

• Skybox AI Quick Start: Building Your First Integration – Blockade Lab

What Services Does Skybox AI Provide?

Skybox AI consists of multiple access methods:

Front End: Web-based interface at skybox.blockadelabs.com

API Services: Programmatic access via api-documentation.blockadelabs.com

Official Integrations: Unity (verified), Node.js/TypeScript, Laravel/PHP SDKs

MCP Client: Model Context Protocol integration for Claude

Community Tools: Unreal Engine plugin (community-maintained)

Why Skybox AI?

Skybox AI generates production-quality 360° environments in seconds from text prompts, eliminating manual environment creation and expensive HDRI asset sourcing. This accelerates development cycles and enables dynamic content generation at scale.

What is a Skybox and Why is it Important?

A skybox is a 360° surrounding environment that creates depth and space in 3D applications. It establishes atmosphere, provides realistic environmental lighting via HDRI, performs better than rendering distant geometry, and makes environments feel expansive without complex assets.

What is Skybox AI and Why is it Useful?

Skybox AI generates 360° equirectangular environments at 8K standard resolution (up to 16K), with exports available in HDRI/EXR for professional rendering, cube maps for game engines, and platform-specific formats like Roblox cube maps.

Applications benefit from dynamic environment generation based on gameplay or user input, rapid prototyping without asset pipelines, infinite variety without storage overhead, and automated workflow integration via API.

What is Skybox AI Good At?

Skybox AI is the leading application for creating equirectangular generative AI HDRI images in 32-bit format. It excels at quickly generating photorealistic and stylized skyboxes for export and integration into 3D game engines including Unity, Unreal Engine, and Roblox. This saves money in production pipelines by creating content from scratch or remastering content for newer engines.

The Remix tool allows users to upload a 360° image or use an existing generation as the basis for new creations. Granular controls modulate the influence applied to the remix process, while a structure retention toggle enables stylized versions of existing images.

What is Skybox AI Not Very Good At?

Skybox AI is not designed for creating humans or animals. It performs poorly when creating spaces without a remix input.

Skybox AI cannot currently create skyboxes that reference other skyboxes to form coherent connected spaces. This capability is under development as a future feature.

Equirectangular video exports are not currently supported. Social media post videos are available for export in portrait, landscape, and square formats.

What Exists Within the Prompt Bar?

The most basic implementation of Skybox AI consists of Text Prompt Input, Style ID from our dropdown list, and a Generate Button.

Prompt Text Input: Provide description of scene

Negative Prompt Toggle: When toggled, input what you do not intend on generating

Enhance Prompt Toggle: When toggled, AI creates a prompt optimized for Skybox AI

Style Dropdown: Choose from over 30 different styles

Generate Button: Request a generation or Cancel (prior to Processing status)

Global Settings in Profile

Important: Global settings override requests made by API.

Strict Prompt: When enabled, affects all generations processed through both Front End and API

Enhance Prompt: When enabled, affects all generations processed through both Front End and API

Custom Watermarking: Upload a custom watermark image to appear on all generations across both Front End and API. Includes preview functionality to view the result before committing.

These settings apply account-wide across all generation methods.

Developer Note: Enhance Prompt and Strict Prompt take granular control away from the user, meaning that achieving complete control of a scene can be difficult. We suggest accompanying this functionality with informational text such as: "Describe the scene you would like to generate - e.g., [example]"

Authentication

API uses x-api-key header. SSO available for Scale tier and above.

Generation Response Times

Standard: Within 45 seconds

Dedicated hardware: 25 seconds

Fastest hardware (on request): 15 seconds

Queuing and Batch Processing

Queuing supported but excessive use causes service disruption. Consult sales for batch processing solutions.

Plan Selection for Development

Individual plans not recommended for multi-user applications (1 concurrent generation only).

Team plans are recommended for application development as they allow for overage pricing and concurrent generation: https://skybox.blockadelabs.com/team-plans 

All include scalable overage, fastest speeds, expedited support, commercial licensing.

Service level agreements are provided on request for customers on Scale plans.

Why Overage Matters

Without overage, applications stop generating at monthly limit. Team Plans include scalable overage pricing.

Why Generation Concurrency Matters

Concurrency determines how many generations can process simultaneously.

Example: With Starter's 2 generations per minute:

• 2 requests arrive → both process immediately
• 3rd request waits for next minute window

Calculating your needs: 50 concurrent users generating every 2 minutes with 30-second generation time = 24 generations per minute needed (requires Scale or higher).

Real-world impact: If you have 1 concurrent generation and 2 customers send requests, the first generation processes (up to 45 seconds), then the second begins. First user receives their image in 45 seconds; second user waits 1.5 minutes total. Proper concurrency ensures positive user experience.

Implications of Ignoring This Advice

Insufficient plans cause service outages, user churn, emergency upgrades, and reputation damage.

Without overage: hard service failure mid-month, emergency support contact for manual intervention, lost revenue during downtime.

Inadequate concurrency: degraded user experience, inability to scale during success or virality, competitive disadvantage.

Custom Events or Unique Requirements

Contact Blockade Labs for:

• High-volume campaigns (conferences, product launches)
• Custom resolution or format requirements
• White-label implementations
• Enterprise SLAs
• Bulk generation discounts

Enterprise solutions available with custom pricing and infrastructure allocation.

Service Status Monitoring

Monitor at status.blockadelabs.com for API availability, queue status, platform-specific services, maintenance, and incidents.

Usage Reports

Team customers receive usage reports upon subscription renewal breaking down: Generation Requests, Errors, Aborts, Success Rate, Complete Generations, and Billing Amount.

Important: Aborted and Error generations are not included in completed generations and are not billed.