Skip to content

Getting Started

Welcome to the vLLM Ascend plugin. This section guides you from environment setup to your first inference workload.

Quick Navigation

  • Quick Start: Run your first model using a prebuilt image.
  • Installation Guide: Set up the driver, CANN, Python, Docker, and source installation environments.
  • Model Tutorials: Find deployment instructions for specific models.
  • Feature Tutorials: Learn about advanced features such as PD disaggregation and context parallelism.
  • FAQ: Troubleshoot common deployment and runtime issues.