> For the complete documentation index, see [llms.txt](https://aiflow-2.gitbook.io/aiflow/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://aiflow-2.gitbook.io/aiflow/platform-structure/layered-architecture/service-layer.md).

# Service Layer

## Service Layer

Purpose: Acts as the execution engine for AIFlow agents, supporting multi-modal ML tasks and secure runtime environments.

Key Features:

* Compatibility with popular AI models (e.g., Nova, Gemini, LLaMA)
* Unified embedding vector processing for text, images, and audio
* gRPC-based remote procedure calls for efficient agent interactions
* Hardware-backed security via Intel SGX or AWS Nitro Enclaves
* Kubernetes-based distributed deployment for large-scale concurrency
* Supports 10,000 concurrent agent runs
* Processes 50,000 inference requests per second