# GPU Infrastructure for 3D Rendering
Pictor Network’s GPU infrastructure for 3D rendering operates under two complementary models:
* **GPU Aggregator Model**\
Functioning as a global GPU aggregator, we connect and manage multiple Render Farms and GPU providers.\
Creators can access the most efficient 3D Rendering services through one platform, with optimized routing for speed, cost, and availability.
* **Decentralized GPU Render Farm Model**\
Beyond the aggregator model, we run our own decentralized render farm powered by decentralized GPU nodes (Worker nodes).\
Each node processes 3D rendering jobs securely and efficiently, contributing to the aggregator network as an independent compute source.
This dual-model approach ensures scalability, transparency, and cost efficiency across all 3D rendering workloads.
## 1. GPU Aggregator Model
### General Model
In this **GPU** **Aggregator Model**, Pictor Network acts as a bridge between Creators and global GPU Render Farms, providing a unified platform to streamline 3D rendering workflows.
Participants in this model include Creators, Pictor Network, Render Farms, and Checker Nodes.
**1. Creators**
Artists, studios, or organizations that require rendering services for their 3D projects. Instead of researching and working with multiple render farms individually, they can use the Pictor Network platform to access various rendering resources in one place. Through Pictor’s single interface, they can submit jobs, monitor progress, review results, and receive final results seamlessly.
**2. Pictor Network**
Pictor Network acts as an aggregator, connecting Creators to global Render Farms. Pictor Network combines the fragmented rendering ecosystem into a single, cohesive network, ensuring compatibility, transparency, and efficiency.
**3. Render Farms**
Render Farms supplies the GPU compute power for 3D jobs submitted via Pictor Network. They provide rendering services, including hardware resources and software resources. By integrating with Pictor Network, they enable Creators to access their resources through a unified platform.
**4. Checker Nodes**
Checker Nodes are the trusted validators of the decentralized GPU network. They verify rendered results, ensuring job accuracy, transparency, and trust for the entire system.
### 3D Rendering Flow
{% stepper %}
{% step %}
**Job Submission**
Creators submit jobs through the Pictor (Web) App.
{% endstep %}
{% step %}
**Cost Calculation**
The Pictor Network protocol calculates the rendering cost based on job complexity and resource needs. This information is attached to a smart contract and sent across the Pictor network.
{% endstep %}
{% step %}
**Task Allocation**
Jobs are split into smaller tasks and distributed to compatible Render Farms based on the most optimal cost and time routes.
{% endstep %}
{% step %}
**Job Processing**
Render Farms process tasks and send regular progress updates to the Pictor App, so Creators can monitor rendering progress in real-time.
{% endstep %}
{% step %}
**Validation**
Checker Nodes validate rendered outputs produced by Render Farms to ensure compliance with job requirements.
{% endstep %}
{% step %}
**Payment and Delivery**
Upon successful validation, rendered outputs are delivered to Creators. Payment is released via smart contract to Render Farms, and details are logged on-chain for transparency.
{% endstep %}
{% endstepper %}
## **2. Decentralized GPU Render Farm Model**
### General Model
In this **decentralized GPU render farm model**, Creators, Workers, Checker Nodes, and the Pictor Network work together to handle rendering job submission, processing, verification, and payments.
Four main participants include:
**1. Creators**
Artists, studios, or anyone who needs high-performance GPU power for their 3D projects. They submit tasks to the network and benefit from the decentralized and cost-effective GPU resources provided by Workers.
**2. Workers**
GPU owners who contribute their computational power to the network for 3D rendering jobs submitted by Creators. Workers can be individuals or organizations with idle GPUs looking to maximize their utilization and monetization.
**3. Checker Nodes**
Checker Nodes play a critical role in maintaining the efficiency, reliability, and quality of the Pictor network. They help verify and validate the render results produced by the Workers.
**4. Pictor Network**
The protocol facilitates the connection and coordination between Creators, Workers, and Checker Nodes, ensuring smooth operation and optimal resource allocation within the Pictor decentralized GPU network.
### 3D Rendering Flow
Pictor Network’s 3D Rendering service works under a SaaS model, supporting almost all 3D software and renderers (render engines) in the 3D graphics industry.
Here’s how the 3D rendering process works:
{% stepper %}
{% step %}
**Job Submission**
Creators submit 3D rendering jobs through the Pictor Web App.
{% endstep %}
{% step %}
**Cost Calculation**
The Pictor protocol calculates the rendering cost based on job complexity and resource needs. This information is attached to a smart contract and sent across the Pictor network.
{% endstep %}
{% step %}
**Job Processing**
The job is assigned to Workers via the Pictor Worker App or Docker Container, which integrates the necessary 3D software and renderers to execute the rendering tasks.
{% endstep %}
{% step %}
**Quality Verification**
Checker Nodes validate rendering outputs to ensure they meet the Creators’ requirements.
{% endstep %}
{% step %}
**Payment**
Upon successful validation, the rendered outputs are delivered to Creators, and payments are distributed to Workers through the smart contract.
{% endstep %}
{% endstepper %}
---
# Agent Instructions: Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter:
```
GET https://docs.pictor.network/docs/company/how-pictor-network-works/gpu-infrastructure-for-3d-rendering.md?ask=
```
The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.