> For the complete documentation index, see [llms.txt](https://docs.grafilab.io/home/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.grafilab.io/home/whitepaper/grafi-co-builder/how-it-works.md).

# How It Works

**Inference API**

1. Connect Your Application: Use the Co-Builder interface to link your application or agent with the Inference API.
2. LLM Selection: Choose the preferred LLM that suits your apps/agents.
3. Real-Time Processing: The API processes data and delivers results with optimized speed and accuracy.
4. Monitor Usage: Track performance metrics and adjust settings to optimize for cost and efficiency.

**Deploy Your Own Apps/Agents**

1. Upload your pre-trained dataset: Developers upload pre-trained dataset or configure new ones within the Co-Builder platform.
2. Select Resources: Choose from centralized or decentralized GPUs for deployment, ensuring scalability and cost efficiency.
3. Train and Scale: Co-Builder handles training and scaling based on performance requirements.
4. Monitor Performance: Use real-time dashboards to oversee training speed, utilization, and outcomes.
5. Monetize: Get listed on the AI App Store for subscription earning rewards for usage.